Chinese herb samples were harvested from Nileke County, Yili, Xinjiang, in August 2012, which were identified as the dry roots of Aconitum soongoricum Stapf. by Yonghe Li (the chief Traditional Chinese Medicine pharmacist in the Fourth Clinical Medical College of Xinjiang Medical University).

The extraction and separation of alkaloids were performed as previously described (11,12). In brief, 10 kg dry roots of Aconitum soongoricum Stapf. were crushed (through a 1x1 inch 20-mesh sieve) and extracted using 95% ethanol (cat. no. 130105; Xi'an Chemical Reagent Factory) at 25˚C and this was repeated 3 times (27 l ethanol each time). The percolation extraction method was used (13). The dichloromethane extract was concentrated by a rotary evaporator at 40˚C to remove the dichloromethane solvent, and the saturated n-butanol extract was concentrated by a rotary evaporator at 50˚C to remove the n-butanol solvent. The extract was concentrated by thin-film evaporation, and then dissolved with 5.5 l 2% HCl and extracted using petroleum ether (a total of 8 l; cat. no. 20131105; Tianjin Hongyan Reagent Factory). The remaining aqueous solution was adjusted to pH=4 with ammonia solution, which was then subjected to the extraction with dichloromethane three times (6 l each time; cat. no. 20131128; Tianjin Chemical Reagent Co., Ltd.). The dichloromethane extract was gathered, and the portion with pH=4 was obtained. The remaining aqueous solution was adjusted to pH=8 with ammonia solution, which was subjected to the extraction with dichloromethane three times (8 l each time), and the dichloromethane phase was obtained (pH=8). The aqueous layer was adjusted to pH=11 with 10% NaOH and extracted using dichloromethane three times (6 l each time) to obtain the portion with pH=11. The aqueous layer was then extracted with saturated n-butanol (2 l), to obtain the n-butanol portion. The portions obtained were separated by the repeated silica gel, Sephadex LH-20 gel column (Amersham Pharmacia Biotech AB) and recrystallization. The fractions were concentrated, and the solvent was removed, prior to the column purification.

The pH=4 portion was eluted with petroleum ether-ethyl acetate-diethylamine (cat. no. 20131213; Tianjin Hongyan Reagent Factory; 8:2:0.5, V/V/V; flow rate, 20 ml/min; 1 l eluate was collected each time), and the eluates from the 18th to 23th washing rounds were collected, gathered, crystallized and filtered, to obtain the Compound 1 (1.5449 g; solid colorless crystal), which was identified as a single compound by the thin layer chromatography (TLC; with the Rf value of 0.2), as previously described (14). The potassium bismuth iodide was used as the chromogenic agent for the TLC method.

The portion of pH=8 was eluted with petroleum ether-ethyl acetate-diethylamine (10:1:0.5; flow rate, 20 ml/min; 1 l elute was collected each time), and the elutes from the 6th to 11th rounds were collected, crystallized and filtered, to obtain Compound 2 (1.3345 g), which was determined to be a single compound according to the TLC (with an Rf value of 0.37).

The pH=8 portion was eluted with petroleum ether-ethyl acetate-diethylamine (10:1:0.5; flow rate, 20 ml/min; 1 l elute was collected each time), and the elutes from the 1st to 3rd rounds were collected, which were then subjected to the Sephadex LH-20 gel column, followed by eluting with chloroform-petroleum ether-methanol (5:5:1; at a flow rate of 5 ml/min). A total of 15 ml elute was collected each time, and the elute from the 7th eluting round was collected, crystallized, filtered and re-crystallized with acetone, to obtain Compound 3 (100 mg), identified as a single compound based on the TLC (with an Rf value of 0.52).

The pH=8 portion was eluted with petroleum ether-ethyl acetate-diethylamine (7:3:0.5; flow rate, 20 ml/min; 1 l elute was collected each time), and the elutes from the 4th to 9th rounds were collected, crystallized and filtered, to obtain Compound 4 (1.2910 g), identified as a single compound by TLC (with an Rf value of 0.6).

The pH=4 portion was eluted with petroleum ether-ethyl acetate-diethylamine (10:1:0.5; flow rate, 20 ml/min; 1 l elute was collected each time), and the elutes from the 3rd to 7th rounds were collected, subjected to the silica gel H column chromatography, followed by the petroleum ether-ethyl acetate-diethylamine (20:1:0.5) elution and pressurization. A total of 50 ml elute was collected each time, and the elute from the 5th round was collected, concentrated, crystallized and filtered, to obtain Compound 5 (103 mg), identified as a single compound by TLC (with an Rf value of 0.82).

The obtained solid substance was subjected to the alkaloid physicochemical identification reaction and melting point (mp) measurement, and the molecular formula of the compound was obtained by ESI-MS as described previously (15). The reports on the diterpene alkaloids were retrieved from the literature, and the structural identification and analysis of the compounds were performed using ¹H-NMR and ¹³C-NMR spectroscopy as described previously (16) (INOVA-600 and 400 model superconducting nuclear magnetic resonance instrument; Varian Medical Systems).

For the chromatographic conditions and system suitability test, the XBridgeTM-C18 column (250x4.6 mm, 5 µm; Waters Corporation) was used, with an octadecylsilane bonded silica filler. The methanol-water-chloroform-triethylamine (volume ratio of 67:33:2:0.1) was used as the mobile phase, and isocratic elution was performed. The conditions were set as follows: The flow rate, 0.8 ml/min; detection wavelength, 235 nm; column temperature, 40˚C; injection volume, 10 µl.

Rheumatoid arthritis HFLS-RA fibroblast-like synoviocytes (suitable for the experiments) (17-19) were derived from Bena Culture Collection (BNCC340356). These cells were cultured using high-glucose Dulbecco's modified Eagle's medium (12800-017; Gibco; Thermo Fisher Scientific, Inc.), containing 10% FBS (FND500; Shanghai ExCell Biology, Inc.), supplemented with 1% penicillin-streptomycin double antibodies (10,000 U; SC30010; Gibco; Thermo Fisher Scientific, Inc.) in a 37˚C, 5% CO₂ incubator, with saturated humidity. The HFLS-RA cells at a confluence of 90% were subjected to the following treatments: i) The blank group, including the normal cultured cells; ii) the lipopolysaccharide (LPS) intervention group, in which the cells were treated with medium containing 100 ng/ml LPS for 26 h; iii) the Leflunomide + LPS intervention group, in which the cells were treated with medium containing 100 ng/ml LPS for 2 h, followed by treatment together with 150 µg/ml leflunomide for another 24 h; iv) the Junggar aconitine + LPS intervention group, in which the cells were treated with medium containing 100 ng/ml LPS for 2 h, followed by treatment together with 350 µg/ml Junggar aconitine for a further 24 h; v) the benzoylaconine + LPS intervention group, in which the cells were treated with medium containing 100 ng/ml LPS for 2 h, followed by treatment together with 1,000 µg/ml benzoylaconine for a further 24 h; and vi) the aconine + LPS intervention group, in which the cells were treated with medium containing 100 ng/ml LPS for 2 h, followed by treatment together with 500 µg/ml aconine for a further 24 h. All treatments were performed at 37˚C.

To prepare the songorine stock solution (prepared by Traditional Chinese Medicine Pharmacy Laboratory and Traditional Chinese Medicine Processing Research Laboratory, Xinjiang Medical University, Urumqi, China ), a total of 100 mg substrate was weighed and dissolved in 500 µl DMSO (D2650; Sigma-Aldrich; Merck KGaA), to obtain a final concentration of 200 mg/ml. To prepare the benzoylaconine stock solution (A0631; Chengdu Must Bio-Technology Co., Ltd.), a total of 20 mg benzoylaconine was weighed and dissolved in 100 µl DMSO, to obtain a final concentration of 200 mg/ml. To prepare the aconitine stock solution (MUST-14012802; Chengdu Must Bio-Technology Co., Ltd.), a total of 100 mg drug was weighed and dissolved in 100 µl DMSO to obtain a final concentration of 200 mg/ml.

Cells were seeded onto the 96-well plate, at a density of 5x10⁴ cells/ml. After 8 days, the culture medium was replaced with 10% CCK-8 solution (Beyotime Institute of Biotechnology), and the cells were incubated at 37˚C for 1 h. Next, the optical density (OD) was read on the xMarkTM microplate reader (Bio-Rad Laboratories, Inc.) at an absorbance of 450 nm, and the growth curve was plotted accordingly. The inhibition rate was calculated using the following formulation: Inhibition rate=(OD_{blank control}-OD_sample)/(OD_{blank control}-OD_{reagent control}) x100%. The IC₅₀ value was calculated by the probit analysis using SPSS 19.0 software (IBM Corp.).

Cells were seeded onto a 96-well plate, at a density of 5x10⁴ cells/ml, and cultured in a 37˚C, 5% CO₂ incubator for 24 h. Following adhering, the cells were incubated with 100 µl songorine at indicated concentrations (0, 100, 300, 500, 700 and 900 µg/ml), 100 µl benzoylaconine at indicated concentrations (0, 500, 1,000, 1,500, 2,000 and 3,000 µg/ml), or 100 µl aconitine at indicated concentrations (0, 100, 500, 1,000, 1,500 and 2,000 µg/ml). After 24, 48 and 72 h, the culture medium was discarded, and the cells were treated with 100 µl 10% CCK-8 solution for at 37˚C for 1.5 h. Next, the OD was read at 450 nm using a microplate reader. Cell images were captured using a fluorescence inverted microscope (magnification, x200; Eclipse TS100-F; Nikon).

The contents of IL-6 (EH004-48; Shanghai ExCell Biology, Inc.), IL-1β (EH001-48; Shanghai ExCell Biology, Inc.), TNF-α (EH009-48l; Shanghai ExCell Biology, Inc.) and PG-E2 (CSB-E07965h; Wuhan Huamei Bioengineering Co., Ltd.) in the culture supernatant were measured using ELISA kits. Cells were seeded onto the 96-well plate at a density of 5x10⁴ cells/ml, and cultured in a 37˚C, 5% CO₂ incubator for 24 h. Drug intervention was performed following cell adhering and, at 2 h before intervention, the cells were treated with 100 ng/ml lipopolysaccharides (LPS; L6529-1MG; Sigma-Aldrich; Merck KGaA) at 37˚C for 24 h. The LPS treatment was to simulate an in vitro inflammatory model and the cells would produce an inflammatory reaction following LPS treatment (20). Next, the cells were treated with 150 µg/ml leflunomide, 350 µg/ml songorine, 1,000 µg/ml benzoylaconine and 500 µg/ml aconitine at 37˚C for 24 h. ELISA was performed with commercially available kits (IL-6 kit, EH004-48, Shanghai ExCell Biology, Inc.; IL-1β kit, EH001-48, Shanghai ExCell Biology, Inc.; TNF-α kit, EH009-48, Shanghai ExCell Biology, Inc.; and PG-E2 kit, CSB-E07965h, CUSABIO), according to the manufacturer's protocols.

Total RNA was extracted from the cells using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.). The cDNA was obtained using the TransScript One-Step gDNA Removal and cDNA Synthesis SuperMix (AT311; TransGen Biotech Co., Ltd.) for 25˚C for 10 min, 42˚C for 30 min and 85˚C for 5 sec. RT-qPCR was performed using the QuantiNava SYBR-Green kit (208054; Kaijie). Primer sequences were as follows: HIF-1α forward, 5'-CACCACAGGACAGTACAGGAT-3' and reverse, 5'-CGTGCTGAATAATACCACTCACA-3'; VEGFA forward, 5'-CAGAAGGAGGAGGGCAGAATC-3' and reverse, 5'-GGTCTCGATTGGATGGCAGT-3'; TLR4 forward, 5'-AGACCTGTCCCTGAACCCTAT-3' and reverse, 5'-CGATGGACTTCTAAACCAGCCA-3'; and β-actin forward, 5'-ACAGAGCCTCGCCTTTGCC-3' and reverse, 5'-GAGGATGCCTCTCTTGCTCTG-3'. The PCR system consisted of 5 µl SYBR Select mix, 0.05 µl Rox, 0.7 µl primer each, 1 µl cDNA and 10 µl ddH₂O. The reaction conditions were as follows: 94˚C for 30 sec; 94˚C for 5 sec, for a total of 40 cycles; followed by 60˚C for 34 sec. Target gene expression levels were calculated using the 2^-ΔΔCT method (21).

Cells were collected and lysed using RIPA lysis buffer (Boster Biological Technology). Protein concentrations were determined using the BSA method. A total of 30 mg protein per lane was separated by 10% SDS-PAGE, which was then electronically transferred onto polyvinylidene difluoride membranes. Following blocking with 5% skimmed milk at room temperature for 1 h, the membrane was incubated with primary antibodies against HIF-1α (1:1,000 dilution; cat. no. ab113642; Abcam), VEGFA (1:1,000 dilution; cat. no. ab1316; Abcam), TLR4 (1:1,000 dilution; cat. no. ab13867; Abcam) and β-actin (1:800 dilution; cat. no. D110024; Sangon Biotech) at 4˚C overnight. Next, the membranes were incubated with the HRP-conjugated pierce goat anti-rabbit IgG (1:10,000 dilution; cat. no. 31460; Thermo Fisher Scientific, Inc.) or pierce goat anti-mouse IgG (1:10,000 dilution; cat. no. 31430; Thermo Fisher Scientific, Inc.) at room temperature for 1 h. The protein bands were detected using the SuperSignal West Prico Chemiluminescent Substrate (34080; Thermo Fisher Scientific, Inc.), and the images were captured and analyzed using Chemi Analysis software.

Data are expressed as the mean ± standard deviation. SPSS 19.0 software (IBM Corp.) was used for statistical analysis. Experiments were performed in triplicates. One-way analysis of variance, followed by Tukey's test, was performed for group comparisons. P<0.05 was considered to indicate a statistically significant difference.

Compound 1 was a colorless solitary crystal. The ¹H-NMR (400 MHz, CDCl3) indicated 1 nitrogen ethyl group (δH: 1.10, 3H, t, J = 7.0 Hz); and 4 methoxy groups (δH: 3.11, 3.27, 3.34, 3.75, s, each 3H; Table I). The chemical shift in the ¹³C-NMR spectroscopy was virtually the same as the features for aconitine according to a previous study (22) (Table II). Based on the ESI/MS, the molecular ion peak m/z was found at 646 [M+H]⁺. According to the nitrogen rule, it was inferred that the compound contained an odd number of nitrogen atoms. The ¹³C-NMR showed 34 carbon signals. Accordingly, the molecular formula was inferred as C₃₄H₄₇NO₁₁, with an unsaturation degree of 12. NMR suggested that it may be a C₁₉-aconitine-type diterpene alkaloid. The compound and aconitine reference substance developed the same color spots at the same positions in various TLC developments. The compound structure is presented in Fig. 1A.

Compound 2 was a colorless solitary crystal. Its ¹H-NMR data are presented in Table I. The chemical shift in the ¹³C-NMR spectroscopy was virtually the same as the features for songorine in a previous study (23) (Table III). Based on the ESI/MS, the molecular ion peak m/z was observed at 358 [M+H]⁺. According to the nitrogen rule, it was inferred that the compound contained an odd number of nitrogen atoms. The ¹³C-NMR showed 22 carbon signals. Accordingly, the molecular formula was inferred as C₂₂H₃₁NO₃, with an unsaturation degree of 8. NMR suggested that it may be a C₂₀-napelline-type diterpene alkaloid. Therefore, the compound was identified as songorine. The compound structure is presented in Fig. 1B.

Compound 3 was a colorless amorphous powder. The ¹H-NMR (400 MHz, CDCl3) indicated 1 nitrogen ethyl group (δH: 1.04, 3H, t, J = 8 Hz), 1 CH group (δH: 3.42, 1H, brs) and 2 CH3 groups (δH: 0.74, 1.38, s, each 3H; Table I). The chemical shift in the ¹³C-NMR spectroscopy was virtually the same as the features for 16, 17-dihydro-12β, 16β-epoxynapelline according to a previous study (24) (Table IV). Based on the ESI/MS, the molecular ion peak m/z was observed at 360 [M+H]⁺. According to the nitrogen rule, it was inferred that the compound contained an odd number of nitrogen atoms. The ¹³C-NMR showed 22 carbon signals. Accordingly, the molecular formula was inferred as C₂₂H₃₃NO, with an unsaturation degree of 7. NMR suggested that it may be a C₂₀-napelline-type diterpene alkaloid. Therefore, the compound was identified as 16, 17-dihydro-12β, 16β-epoxynapelline. The compound structure is presented in Fig. 1C.

Compound 4 was a colorless amorphous powder. The ¹H-NMR (400 MHz, CDCl3) indicated 1 nitrogen ethyl group (δH: 1.09, 3H, t, J=4 Hz), 1 CH group (δH: 3.91, 1H, brs) and 1 CH3 group (δH: 0.77, s, 3H; Table I). The chemical shift in the ¹³C-NMR spectroscopy was virtually the same as the features for 12-epi-napelline according to a previous study (25) (Table V). Based on the ESI/MS, the molecular ion peak m/z was observed at 360 [M+H]⁺. According to the nitrogen rule, it was inferred that the compound contained an odd number of nitrogen atoms. The ¹³C-NMR showed 22 carbon signals. Accordingly, the molecular formula was inferred as C₂₂H₃₃NO, with an unsaturation degree of 7. NMR suggested that it may be a C₂₀-napelline-type diterpene alkaloid. Therefore, the compound was identified as 12-epi-napelline. The compound structure is presented in Fig. 1D.

Compound 5 was a colorless amorphous powder, with an mp of 169-170˚C, C₃₄H₄₇NO₁₀, ESI/MS ([M+H]⁺, m/z 630), ¹H-NMR (500 MHz, CDCl₃) δppm: 1.09 (3 H each, t, J=7 Hz, NCH₂CH₃), 1.39 (3 H each, s, COCH₃), 3.18, 3.29, 3.31, 3.76 (3 H each, s, 4x-OCH₃), 4.39 (1 H each, d, J=3 Hz, 15-αOH), 4.48 (1 H each, dd, J=3, 5.5 Hz, 15-βH), 4.89 (1 H each, d, J=5 Hz, 14-βH), 7.42-8.08 (5 H each, m, Ar-H; Table I).

The aforementioned data, as well as the results from the ¹³C-NMR spectroscopy, were virtually the same as the features for deoxyaconitine, according to a previous study (24) (Table VI). Based on the ESI/MS, the molecular ion peak m/z was observed at 630 [M+H]⁺. According to the nitrogen rule, it was inferred that the compound contained an odd number of nitrogen atoms. The ¹³C-NMR showed 34 carbon signals. Accordingly, the molecular formula was inferred as C₃₄H₄₇NO₁₀, with an unsaturation degree of 12. NMR suggested that it may be a C₂₀-aconitine-type diterpene alkaloid. Therefore, the compound was identified as deoxyaconitine. The compound structure is presented in Fig. 1E.

The proliferation of HFLA-RA cells was assessed using the CCK-8 assay, and a growth curve was obtained. At 24 h after seeding, the cells were in the latency phase, and at 48 h they were in the logarithmic growth phase. Starting from the 4th day, the growth platform began and it lasted until day 7 (Fig. 2). The effects of drug interventions on the growth of HFLS-RA cells were then assessed (Fig. 3A-D). The results of the present study demonstrated that, for the effects of songorine on the cell proliferation, the IC₅₀ values for the intervention for 24, 48 and 72 h were 491.4, 436.7 and 385.6 µg/ml, respectively (Fig. 3E). Furthermore, the IC₅₀ values for the benzoylaconine intervention for 24, 48 and 72 h were 1,632.0, 1,552.6 and 1,332.5 µg/ml, respectively (Fig. 3E). Furthermore, the IC₅₀ values for the aconitine intervention for 24, 48 and 72 h were 775.1, 679.9 and 609.9 µg/ml, respectively (Fig. 3E). These results suggested that songorine, benzoylaconine and aconitine at different concentrations may inhibit the proliferation of HFLS-RA cells, to different extents.

The effects of drug interventions on the cytokine contents in culture supernatant were subsequently investigated. The results demonstrated that, compared with the blank group, the contents of IL-6, IL-1β, TNF-α and PGE-2 in the culture supernatant were significantly increased by treatment with LPS (P<0.05). Furthermore, compared with the LPS group, the contents of IL-6, IL-1β, TNF-α and PGE-2 in the culture supernatant were significantly declined in the leflunomide + LPS and the drug intervention + LPS groups (P<0.01). Furthermore, the contents of IL-6, IL-1β, TNF-α and PGE-2 in the culture supernatant in the intervention + LPS groups were higher than that in the leflunomide + LPS group (P<0.01; Table VII). These results suggested that, the drug interventions may significantly improve the LPS-induced cellular inflammatory responses.

The effects of drug interventions on the mRNA and protein expression levels of HIF-1α, VEGFA and TLR4 were investigated by RT-qPCR and Western blot analysis, respectively. The results from the RT-qPCR and Western blot analysis demonstrated that, compared with the blank group, the mRNA and protein expression levels of HIF-1α, VEGFA and TLR4 in the cells were significantly increased in the LPS group (P<0.05). Furthermore, compared with the LPS group, all the mRNA and protein expression levels of HIF-1α, VEGFA and TLR4 in the cells were significantly decreased in the leflunomide + LPS and intervention + LPS groups (P<0.01). Furthermore, compared with the leflunomide + LPS group, the mRNA expression levels of HIF-1α, VEGFA and TLR4 were higher in the intervention + LPS groups (P<0.01; Fig. 4; Table VIII).