rBmpA was produced in Escherichia coli BL21 (GE Healthcare, Chicago, IL, USA) using the bacterial expression vector pGEX-6P1 (GE Healthcare) and the following primers with EcoRI and XhoI restriction sites: Forward, 5′-ACGAATTCATGAATAAAATATTGTTGTTGA-3′ and reverse, 5′-AGCTCGTAAATAAATTCTTTAAGAAA-3′ (4,5). Pure ISOF was provided by Dr Weimin Yang [School of Pharmaceutical Science and Yunnan Key Laboratory of Pharmacology for Natural Products, Kunming Medical University (KMU), Kunming, China] and reconstituted at 1×10⁵ µM/ml in dimethyl sulfoxide (DMSO) for stock solution. LPS and phorbol-12-myristate-13-acetate (PMA) were purchased from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany) and ionomycin was purchased from Apollo Scientific Ltd. (Stockport, UK). Recombinant human (rh)IL-4, rhTNF-α and rh-granulocyte macrophage colony-stimulating factor (rhGM-CSF) were purchased from R&D Systems, Inc. (Minneapolis, MN, USA).

The murine macrophage cell line RAW264.7 and human monocytic leukemia cell line THP-1 were obtained from the Kunming Institute of Zoology, Chinese Academy of Sciences (Kunming, China). All cell culture medium, serum and antibiotics were purchased from Gibco; Thermo Fisher Scientific, Inc. (Waltham, MA, USA). All cells were cultured at 37°C with a humidified atmosphere of 5% CO₂. Cellular supernatants were collected for ELISA. Cell lysates were prepared using RNAiso Plus reagent (Takara Bio, Inc., Otsu, Japan).

For RAW264.7 cells, Dulbecco's modified Eagle's medium (high glucose) supplemented with 10% FBS and 1% penicillin-streptomycin was used as culture medium. RAW264.7 cells were seeded in 96-well microplates at a concentration of 1×10⁵ cells/ml and stimulated with 20 µg/ml rBmpA or 20 µg/ml rBmpA+100 µmol/ml ISOF (n=4 wells per condition) for 24 and 48 h. Additionally, a group of cells was cultured in 10% FBS/DMEM medium alone as a blank control and subgroups of these cells were stimulated with 0.1% DMSO (as a solvent control) or 1 µg/ml LPS (as a positive control).

For THP-1 cells, RPMI-1640 supplemented with 10% FBS and 1% penicillin-streptomycin was used as cell culture. THP-1 cells at 5×10⁵ cells/ml were differentiated into macrophages by 100 ng/ml PMA pretreatment for 24 h (10). THP-1 cells at 2×10⁵ cells/ml were pretreated with 200 ng/ml rhIL-4, 100 ng/ml rhGM-CSF, 10 ng/ml rhTNF-α and 200 ng/ml ionomycin for 2 days for differentiation into mature dendritic cells (11). Subsequently, the macrophages and dendritic cells were cultured in 10% FBS/RPMI-1640 and stimulated with 20 µg/ml rBmpA, 20 µg/ml rBmpA + 100 µmol/ml ISOF, 0.1% DMSO (solvent control) or 1 µg/ml LPS (positive control) (n=4 wells per condition) for 24, 48 and 72 h. Additionally, a group of cells was maintained in RPMI 1640 supplemented with 10% FBS medium as a blank control.

The concentrations of TNF-α and IL-6 in the cellular supernatant samples were measured using Mouse TNF-α ELISA kit (cat. no. DKW12-2720-096; Dakawe, Shenzhen, China), Mouse IL-6 ELISA kit (cat. no. DKW12-2720-096; Dakawe), Human TNF-α ELISA kit (cat. no. ELH-TNFα-001; Ray Biotech, Inc., Norcross, GA, USA) and Human IL-6 ELISA kit (cat. no. ELH-IL6-001; Ray Biotech, Inc.) were used according to the manufacturers' protocols.

TNF-α and IL-6 mRNA expression was assessed by RT-qPCR. Total RNA was extracted from cells using the RNAiso Plus reagent, and a PrimeScript Reverse Transcription reagent kit with gDNA Eraser (Takara Bio, Inc.) was used to synthesize cDNA, according to the manufacturer's protocol. mRNA expression was detected using the standard SYBR-Green RT-PCR kit (Takara Bio, Inc.) according to the manufacturer's protocol. The sequences of the specific primers used to amplify murine and human TNF-α, IL-6 and β-actin were as follows: For murine TNF-α, forward, 5′-GGTCCCCAAAGGGATGAGAA-3′ and reverse, 5′-TGAGGGTCTGGGCCATAGAA-3′; for murine IL-6, forward, 5′-TCGGAGGCTTAATTACACATGTTC-3′ and reverse, 5′-CATACAATCAGAATTGCCATTGC-3′; for murine β-actin, forward, 5′-TGCCGCATCCTCTTCCTC-3′ and reverse, 5′-CGCCTTCACCGTTCCAGT-3′; for human TNF-α, forward, 5′-CCTCTCTCTAATCAGCCCTCTG-3′ and reverse, 5′-GAGGACCTGGGAGTAGATGAG-3′; for human IL-6, forward, 5′-ACTCACCTCTTCAGAACGAATTG-3′ and reverse, 5′-CCATCTTTGGAAGGTTCAGGTTG-3′; and for human β-actin, forward, 5′-CAAGGCCAACCGCGAGAAGA-3′ and reverse, 5′-GGATAGCACAGCCTGGATAG-3′ (12). qPCR was performed using SYBR Premix Ex Taq (Takara Bio, Inc.), according to the manufacturer's protocol. The reaction conditions were 95°C for 10 min, and 40 cycles of denaturation at 95°C for 15 sec and annealing/elongation at 60°C for 30 sec. The relative expression was analyzed by the 2^−ΔΔCq method (12).

A total of 45 female Kunming mice (age, 6–8 weeks; weight, 18–22 g) were acquired from the Experimental Animal Center of KMU, Kunming, China. All animals were cared for according to the Guide for the Care and Use of Experimental Animals formulated by the Chinese Council on Animal Care and the Principles of Laboratory Animal Care formulated by Yunnan Provincial Council. The experimental protocol received approval from the Animal Use and Care Committee of KMU and was performed according to Public Health Service Policy on Humane Care and Use of Laboratory Animals by the National Institute of Health (13). The mice were kept in standardized units with filtered air at a controlled temperature of 18–22°C and humidity of 50–60% with a l2 h light/dark cycle and free access to food and water. Prior to experimental procedures, the mice were adapted to the environment for 1 week. The surgical procedures performed in animals were under general anesthesia using 360 mg/kg chloral hydrate [intraperitoneal (i.p.)].

The preparation of ISOF stock and purified rBmpA was performed using the aforementioned methods. Phosphate-buffered saline (PBS) was purchased from Thermo Fisher Scientific, Inc., and diluted to 0.01 M with ultrapure water. The 0.01 M PBS was then used to dilute ISOF stock (1×10⁵ Mm/ml) and purified rBmpA to the concentration of 0.05 mg/ml.

Mice were randomly divided into a PBS control group (n=15), Lyme arthritis group (n=15) and ISOF group (n=15). The Lyme arthritis model was established in the Lyme arthritis and ISOF groups as follows: 50 µl rBmpA (0.05 mg/ml) was injected into the tibiotarsal joint cavity of both hind legs once every 3 days for a total of four injections. PBS (0.01 M) or ISOF solution (1 mg/kg weight, i.p.) was then administered to the mice in the Lyme arthritis and ISOF groups, respectively. In the control group, 0.01 M PBS was administered instead of rBmpA and ISOF.

To assess the extent of arthritis, two researchers blinded to the experimental conditions scored the tibiotarsal joints twice a week following the initial injection. The joints were evaluated according to the following arthritis scoring scale: 0, no joint swelling or redness; 1, slight reddening of the joint but no swelling; 2, slight reddening and swelling of the joint; 3, moderate reddening and swelling of the joint; 4, notable swelling and dysfunction of the joint. The MAI in each group was equal to the sum of arthritis scores divided by the number of mice hind legs. At 4 weeks after discontinuation of injections, dental radiography units were used to obtain X-ray images.

All 45 mice were anaesthetized with 360 mg/kg chloral hydrate (i.p.) and euthanized 4 weeks following the last injection in order to collect the bilateral knee joint, the tibiotarsal joint and the entire paw. The joint tissue was fixed in 4% paraformaldehyde at room temperature for 24 h. Following decalcification with EDTA, all samples were cut into sections 5 µm thick and stained using hematoxylin and eosin at room temperature. Stained sections were observed and photographed with a light inverted microscope.

Data were expressed as the mean ± standard error of the mean. Differences between the mean values of the groups were evaluated by one-way analysis of variance followed by a Student-Newman-Keuls test using the statistical analysis software GraphPad Prism 6.0 (GraphPad Software, Inc., La Jolla, CA, USA). P<0.05 was considered to indicate a statistically significant difference.

The TNF-α concentrations in the cellular supernatants of the different cell groups were detected. TNF-α was quantified at different culture time points using specific ELISA kits. In the supernatant from murine macrophages at 24 and 48 h, it was observed that treatment with LPS or rBmpA significantly increased the level of TNF-α when compared with the blank controls (all P<0.01; Fig. 1A). In turn, ISOF treatment significantly reduced the secretion of TNF-α induced by rBmpA (all P<0.01; Fig. 1A).

Similarly, LPS or rBmpA treatment induced a significant increase in TNF-α in the supernatant of human macrophages at 24, 48 and 72 h compared with the blank control group (all P<0.01; Fig. 1B). Additionally, treatment with ISOF significantly reduced the secretion of TNF-α induced by rBmpA at 24, 48 and 72 h compared with the rBmpA group (all P<0.01; Fig. 1B).

Furthermore, as depicted in Fig. 1C, rBmpA upregulated the levels of TNF-α in the supernatants of human dendritic cells after 24, 48 and 72 h, and LPS after 72 h compared with the blank controls (all P<0.01). Meanwhile, ISOF treatment significantly suppressed the rBmpA-induced secretion of TNF-α at 24, 48 and 72 h (P<0.01).

As for TNF-α, ELISA was used to quantify the supernatant concentrations of IL-6 at different culture time points. For murine macrophages, compared with the blank control group, the LPS or rBmpA groups exhibited significantly increased supernatant levels of IL-6 at 24 and 48 h (P<0.01; Fig. 2A); In turn, ISOF treatment reduced the secretion of IL-6 induced by rBmpA, which was deemed to be significant at 48 h (P<0.01) but not at 24 h (Fig. 2A).

Accordingly, IL-6 levels were significantly increased in the supernatants of human macrophages by LPS or rBmpA compared with the blank controls at 24, 48 and 72 h (P<0.01). Meanwhile, ISOF treatment significantly reduced the secretion of IL-6 induced by rBmpA at 24, 48 and 72 h (P<0.01; Fig. 2B).

For human dendritic cells, the supernatant levels of IL-6were significantly increased by the LPS and rBmpA at 48 and 72 h, compared with the blank controls (P<0.01). Treatment with ISOF significantly reduced the rBmpA-induced increase in IL-6 at 48 and 72 h (P<0.01 and P<0.05, respectively; Fig. 2C).

TNF-α mRNA levels in murine macrophages, human macrophages and human dendritic cells were determined by RT-qPCR. Murine TNF-α expression was significantly increased by LPS or rBmpA at 48 h compared with the blank control group (P<0.01; Fig. 3A). ISOF downregulates murine TNF-α expression induced by the rBmpA at 24 and 48 h (P<0.01). For human macrophages and human dendritic cells, the expression levels of TNF-α were significantly elevated in the LPS and rBmpA groups compared with the blank controls at 24, 48 and 72 h (all P<0.01; Fig. 3B and C). While ISOF treatment significantly reduced the rBmpA-induced upregulation of TNF-α in in human macrophages at 24 h (P<0.01; Fig. 3B), and in human dendritic cells at 48 and 72 h (P<0.01 and P<0.05, respectively).

The IL-6 mRNA levels in murine macrophages, human macrophages and human dendritic cells were determined by RT-qPCR. The results indicated that IL-6 expression levels in murine macrophages, human macrophages and human dendritic cells were significantly increased by LPS or rBmpA treatment when compared with the blank controls at all time points (all P<0.01; Fig. 4). While ISOF treatment reduced the expression of IL-6 induced by rBmpA in human macrophages at 24 and 72 h (both P<0.01; Fig. 4B) but not at 48 h. ISOF treatment also significantly decreased the mRNA expression of IL-6 in rBmpA-stimulated human dendritic cells, which was deemed to be significant at 48 h (P<0.01) but not at 24 or 72 h (Fig. 4C).

On initiation of animal experimentation, mice in the control and ISOF groups exhibited similar feeding behavior and activity levels. At 1–2 days following the initial injection, the mice in the ISOF group presented with slight redness and swelling at the injection site, though this diminished after 3–4 days. The groups administered with rBmpA (Lyme arthritis group and ISOF group) presented slight redness and swelling in the tibiotarsal joints at day 4 after receiving the initial rBmpA injection. The symptoms peaked at days 14–18 and gradually resolved thereafter. A total of 3 mice in rBmpA group with severe inflammation exhibited deformities in the tibiotarsal joint and limited bending ability. The injection of rBmpA caused a significant increase in the MAI compared to that in the control group (all P<0.01; Fig. 5). The treatment of ISOF significantly decreased the MAI compared to rBmpA group (all P<0.01; Fig. 5).

In X-ray images (Fig. 6), the mice with experimental arthritis exhibited swelling of the soft tissue and joint, joint deformity, stiffness and narrow joint spaces. Meanwhile, the mice administered with ISOF exhibited no stiffness and less swelling in soft tissue.

On histopathological examination of control group tissues (Fig. 6G), a regular joint space and orderly arrangement of cells in the synovial membrane were observed. Additionally, the articular synovial surface was neat and there was no infiltration of inflammatory cells. For the arthritis group (Fig. 6H), the joint tissue sections exhibited narrowing of the joint space, increased synovial cells, thickened synovial tissues and obvious fibrous tissue hyperplasia when compared with the control tissues. Furthermore, a high number of inflammatory cells (including polymorphonuclear leukocytes, monocytes-macrophages and lymphocytes) had infiltrated and the synovial membrane and pannus had invaded the joint, causing joint structural damage between bones. For the ISOF group (Fig. 6I), the joint tissue sections exhibited normalization of the joint space and neater rows of joint synovial cells compared with the arthritis group tissues. While the synovial cells remained slightly thickened compared with control tissues, the articular surface appeared neat and no notable inflammatory cell infiltration was observed.