2.1 Animals

Female Wistar rats (160–180 g) were obtained from Shanghai Slac Laboratory Animal Co., Ltd. (Shanghai, China). These animals were housed in specific pathogen-free (SPF) conditions bred with water and food accessible. The animal experiments were carried out in accordance with the protocol by the Experimental Committee of Nanjing Normal University (IACUC-20201203).

2.2 Preparation of CIA animal model and treatment with Phil

To establish the rat CIA model, we followed the previously described protocol (Qiu et al., 2018), which involved an intradermal injection of a 1:1 emulsion of chicken CⅡ (1 mg/ml, Sigma, Saint Louis, MO, USA) and CFA containing Mycobacterium butyricum (Sigma, Saint Louis, MO, USA) in Wistar rats. On day 7, CⅡ (0.5 mg/ml) emulsified with IFA (Sigma, Saint Louis, MO, USA) were subcutaneously boosted in these animals. The rats were immunized with 0.9% saline in the same way as healthy controls. In this study, a total of 42 rats were purchased, with 7 used as healthy controls and the remaining 35 used for CIA modeling. Our laboratory has achieved an approximate 80% success rate in constructing CIA models. Specifically, in this study, 28 out of 35 rats were successfully modeled for CIA. Once reaching an arthritis index score ≥ 2, the 28 CIA rats were randomly divided into various groups (n = 7): vehicle-treated CIA group (CIA rats treated with 0.9% saline); MTX-treated group (CIA rats treated with 3 mg·kg⁻¹ MTX); Phil-treated groups (CIA rats treated with 10 and 20 mg·kg⁻¹ Phil, respectively). Notably, Phil or 0.9% saline was intraperitoneally given once a day for up to 14 days. Phil (HPLC ≥ 98%, C₂₇H₃₄O₁₁, MW: 534.55, Catalog #: L-010) was purchased from Must Bio-Technology Co., Ltd. (Sichuan, China). Its dosage was determined based on our preliminary experiments and previous reports (Qu et al., 2010; Zhong et al., 2013; Cheng et al., 2017; Wang et al., 2019). We used Methotrexate (MTX) as a positive drug, which was obtained from Shanghai Sine Pharmaceutical Co., Ltd. (Shanghai, China) and its dosage was determined according to clinical routine usage.

2.3 Clinical assessment

To evaluate the severity of arthritis, clinical scores were daily detected by two independent and trained observers from the day of the second immunization. Arthritis index score was obtained on a scale of 0 to 4 for each limb. The detailed scoring standard was described in our previous reports (Qiu et al., 2018). The two hind limbs of rats were evaluated, resulting in a maximal score of 8 per animal. Simultaneously, the change in paw volume was determined by a volume displacement plethysmometer. In addition, the rats were daily weighed and the change in body weight was analyzed.

2.4 Measurement of serum biochemical parameters

At the end of the treatment experiment, the rats were euthanized, and their serum was collected to measure the levels of liver damage markers (AST, ALT) and indicators of renal function (Cre, BUN) using commercial kits from Jiancheng Bioengineering Institute (Nanjing, Jiangsu, China), following the manufacturer's instructions.

2.5 Radiographic assessment

To evaluate bone destruction in CIA rats, the left hind ankles and paws were collected for radiographic analysis. The degree of bone destruction was assessed using a scale of 0 to 3, based on our previously described method (Qiu et al., 2018; Liu et al., 2018). Radiographic scoring was conducted in a double-blind manner to minimize bias.

2.6 Histopathological evaluation

After radiographic analysis, the left ankles were immersed in an EDTA solution to remove calcium. Once the samples were soft enough to be penetrated by a needle without resistance, they were dehydrated, embedded, and sectioned into 4 µm thickness. The sections were stained with haematoxylin and eosin (H&E) and evaluated for histopathological changes including synovial hyperplasia, pannus formation, cartilage erosion, and bone destruction using a semi-quantitative scoring system based on our previous reports (Qiu et al., 2018; Liu et al., 2018).

2.7 Assessment of proinflammatory factors in joints

The right hind ankles and paws of rats were homogenized with cold phosphate buffer solution, as described in our previous reports (Qiu et al., 2018; Yu et al., 2021). TNFα and IL-1β in the joint homogenates were determined using the ELISA kits (SenBeiJia Bio-Tech, Nanjing, Jiangsu, China).

2.8 Cell culture

To obtain FLS cells, we isolated synovial tissues from the knee joints of CIA-Vehicle rats and digested them using a double enzymatic method with 0.25% trypsin and 0.4% type II collagenase, following the protocol described in our previous report (M. Liu et al., 2018; Z. Liu et al., 2018). The resulting cells were cultured in H-DMEM medium (Gibco, Grand Island, NY, USA) under standard conditions of 37 °C and 5% CO₂. In vitro experiments were conducted using cells from passages three to nine.

2.9 Cell viability assay

The FLS cells were incubated with Phil at concentrations ranging from 0 to 640 µM for 48 h, followed by the addition of MTS/PMS (Sigma, Saint Louis, MO, USA) and further incubation for 4 h. The absorbance was measured using a microplate reader at a wavelength of 490 nm.

2.10 Cell migration assay

When the cell monolayers reached 80% confluency, a sterile 200 µl pipette tip was used to create scratches. The cells were washed with PBS and treated with different concentrations of Phil for 4 h, then stimulated with 50 ng∙ml⁻¹ TNFα (Peprotech, Rocky Hill, NJ) for 24 h. The migration ratio of the cells was measured using Image J software (NIH, Bethesda, MD, United States).

2.11 Cell proliferation assay

The cells were pre-treated with 5 μM Phil for 4 h, stimulated with 50 ng∙ml^-1 TNFα for 24 h, and subsequently incubated with 10 μM EdU (KeyGen Biotech, Nanjing, Jiangsu, China) for 6 h. After fixation with methanol, the nuclei were stained with Hoechst 33342. The cell proliferation rate was determined by calculating the number of EdU-positive cells relative to the total number of Hoechst-positive cells.

2.12 Cell apoptosis assay

Flow cytometry was used to assess cell apoptosis. After treatment with different concentrations of Phil for 24 h, Annexin V-FITC and propidium iodide (PI) staining solutions (KeyGen Biotech, Nanjing, Jiangsu, China) were added to the cells, followed by analysis using flow cytometry (FACScan, Becton Dickinson).

2.13 Quantitative real-time PCR

The FLS cells were treated with different doses of Phil for 1 h and incubated with 50 ng·ml⁻¹ TNFα for 24 h. TRIzol reagent (Invitrogen, Carlsbad, CA, USA) was added to the wells and total RNA was extracted. After reverse transcription with purified RNA and OligdT primers, real-time PCR was conducted using SYBR-Green qPCR Master Mixes (Vazyme Biotech, Jiangsu, China). The specific PCR primer pairs for TNFα, IL-6, IL-1β, MMP2, MMP9 and β-actin were utilized, as described previously (Qiu et al., 2018; Yu et al., 2021).

2.14 Western blot analysis

The FLSs were treated with different doses of Phil and then stimulated with 50 ng∙ml⁻¹ TNFα for 15 min. Radioimmunoprecipitation assay buffer was added to the wells and the supernatants were collected. The protein in the supernatants was separated using SDS-PAGE and transferred onto Polyvinylidene Fluoride membranes (PVDF, Millipore, Darmstadt, Germany). These PVDF membranes were blocked and then incubated with antibodies against p-p38, p-JNK, p-ERK, p38, JNK, ERK (Cell Signaling Technology, Beverly, MA, USA), IκBα and GAPDH (Santa Cruz Biotechnology, Dallas, CA, USA) overnight at 4 °C. Corresponding horseradish peroxidase-conjugated antibodies were used to incubate the membranes, and ECL detection reagent was added to visualize the bands. The relative expression was analyzed using Image J software.

2.15 Immunofluorescent staining

The cells were cultured and seeded onto glass cover slips in 24-well plates. Following that, they were fixed using methanol and permeabilized with 0.5% Triton-X 100. Subsequently, the cells were incubated with p65 antibody (Cell Signaling Technology, Beverly, MA, USA) and DAPI. Finally, the localization of p65 was observed using a Nikon A1R resonance scanning confocal microscope with a spectral detector (Nikon, Tokyo, Japan).

2.16 Statistics

The experiments were repeated at least three times unless otherwise indicated. For in vivo experiments, t-test was employed to compare the healthy control group and the CIA-Vehicle group, and one-way analysis of variance (ANOVA) was used to compare different treatment groups, followed by Tukey’s post hoc analysis. The data are presented as the mean ± SEM. For in vitro experiments, t-tests were used to compare the TNFα-untreated group and the TNFα-treated but Phil-untreated group, and one-way ANOVA was used to compare different Phil treatment groups, followed by Tukey’s post hoc analysis. The data are presented as the mean ± SD. P < 0.05 was considered statistically significant.

3.1 Synovitis and bone destruction of CIA rats were significantly attenuated by Phil treatment

To assess the therapeutic effect of Phil on RA, a rat model of CIA was established and different concentrations of Phil (10 and 20 mg·kg⁻¹) were intraperitoneally administrated daily for a period of 14 days (Fig. 1). The vehicle-treated group exhibited continuously increasing arthritis scores throughout the treatment period, with a maximum score of 8 (Fig. 2A and B). In contrast, Phil or MTX treatment effectively mitigated the development and progression of CIA (Fig. 2A and B). The changes in paw volume also reflected the efficacy of Phil or MTX in reducing arthritis symptoms (Fig. 2C). Notably, Phil exhibited greater anti-inflammatory potential than MTX, as evidenced by lower scores and paw volumes in Phil-treated CIA rats (Fig. 2A-C).

Close

Fig. 1

The scheme of CIA induction and Phil’s treatment. The rats received their first immunization on day 0, followed by a second immunization on day 7. Once their clinical score reached ≥ 2, CIA rats were randomly assigned to intraperitoneally receive either 10 mg·kg⁻¹ or 20 mg·kg⁻¹ of Phillyrin or vehicle (0.9% saline) daily for 14 days. Additionally, the positive drug group was given 3 mg·kg⁻¹ of MTX once every 3 days. After 14 days of treatment, the rats were sacrificed and their tissues were collected for further analysis.

Export to PPT

Close

Fig. 2

Phil significantly improves the symptoms in the rats with CIA. (A) At the end of the treatment, photographs of the hind paws of the rats with CIA were taken. Hind paw score (B) and Hind paw volume (C) were assessed during the treatment period (mean ± SEM, n = 7, *P < 0.05, **P < 0.01 and ***P < 0.001 versus CIA-Vehicle group); Representative radiographs of hind paws from different groups were shown in (D), and (E) the radiological scoring scale was used to score the different groups (mean ± SEM, n = 7, *P < 0.05, **P < 0.01 and ***P < 0.001).

Export to PPT

Bone destruction, a common feature of CIA animals, was analyzed by X-ray. As shown in Fig. 2D, compared with the normal controls, the hind paws of the vehicle-treated rats showed narrowed joint spacing, severe bone erosion, and rough bone surface. In contrast, Phil treatment significantly and dose-dependently inhibited bone destruction in CIA rats (Fig. 2D and 2E), indicating its potential anti-bone erosive action in RA.

To further investigate the inhibitory effect of Phil on arthritis severity, we performed a semi-quantitative histopathological assessment. Pathological features of RA, such as synovitis, pannus, and bone erosion, were clearly observed in the vehicle-treated CIA group. However, treatment with Phil effectively suppressed these pathological changes, resulting in significantly reduced histopathological scores (Fig. 3A and B).

Close

Fig. 3

Phil significantly ameliorates the pathological changes in the rats with CIA. (A) Representative H&E staining of ankle joints in different groups. (B) Histopathological evaluation of ankle joints in different groups. (mean ± SEM, n = 7, *P < 0.05, **P < 0.01 and ***P < 0.001). B, bone; C, cartilage; J, joint space; P, pannus; S, synovium.

Export to PPT

A preliminary investigation was conducted to explore the potential side effects of Phil on CIA rats (Table 1). As expected, the body weight of the vehicle-treated CIA rats was significantly reduced. However, Phil treatment significantly and dose-dependently suppressed the loss of weight body. In addition, the liver and spleen indexes in the vehicle-treated group were significantly elevated, and this increase was inhibited by MTX or Phil treatment. No significant statistical differences were observed among the normal controls, vehicle-treated, MTX- or Phil-treated groups in terms of heart and kidney indexes, as well as levels of ALT, AST, Cre, and BUN. These preliminary data suggest that the doses of Phil employed in this study are unlikely to cause significant adverse side effects.

3.2 Production of pro-inflammatory cytokines in CIA animals was effectively reduced by Phil treatment

Inflammation is considered to be a key driver of RA pathogenesis. Considerable pro-inflammatory mediators, especially TNFα and IL-1β, are released and play pivotal roles in regulating the initiation and maintainment of joint inflammation (Vanderborgh et al., 2004; Polzer et al., 2010). Thus, the production of TNFα and IL-1β in the rat joint tissues was measured. In comparison to the normal controls, the vehicle-treated CIA rats had markedly increased levels of TNFα and IL-1β. However, these increases were significantly inhibited by Phil treatment in a dose-dependent manner (Fig. 4). These data suggested that anti-arthritic action of Phil might attribute, at least in part, to the decreased expression of pro-inflammatory factors.

Close

Fig. 4

Phil significantly decreases the production of inflammatory cytokines (TNFα and IL-1β) in the joint tissues of the rats with CIA. After homogenizing the right hind ankles and paws of CIA rats, the levels of TNFα (A) and IL-1β (B) were quantified (means ± SEM, n = 7, **P < 0.01 and ***P < 0.001).

Export to PPT

3.3 Proliferation and migration of TNFα-induced arthritic FLSs were significantly suppressed by Phil treatment

To determine the possible cytotoxic effect of Phil in FLSs, an MTS assay was conducted. The result showed that Phil had little influence on the cell viability within the dose range of 0 ∼ 640 μM (Fig. 5A). In this study, no more than 10 μM Phil was used for in vitro experiments.

Close

Fig. 5

Phil treatment suppresses the proliferation and migration of arthritic FLS cells stimulated by TNFα. (A) MTS assay proved that Phil at given doses had little effect on the arthritic FLSs’ viability (means ± SD, n = 3). (B, C) EdU incorporation assay proved that 5 μM Phil could reduce the proliferation rate of arthritic FLSs. Data are presented as means ± SD of three independent experiments (**P < 0.01). (D, E) Representative graphs and statistical analysis of the wound healing assay (means ± SD, three independent experiments, *P < 0.05, **P < 0.01 and ***P < 0.001). (F) Representative graphs and statistical analysis of the apoptosis assay (means ± SD, three independent experiments).

Export to PPT

As is well known, the abnormal hyperplasia of synovial tissues in RA patients is caused by excessive proliferation of over-activated FLSs. Therefore, inhibiting the proliferation of FLSs is considered as a promising strategy for the treatment of RA. In our study, the EdU incorporation results suggested that the proliferation rate of FLSs was markedly enhanced upon TNFα stimulation. But the enhancement was effectively suppressed by 5 μM Phil (Fig. 5B and 5C).

Migration of FLSs to cartilage and bone is a critical step in the deterioration of RA, therefore, suppressing FLSs’ migration is an important approach to prevent the destructive progress of this disease. Our wound healing results showed that Phil could significantly inhibit TNFα-stimulated migration at concentrations greater than 1 μM (Fig. 5D and E). Additionally, we also examined the apoptotic inducing effect of Phil using flow cytometry and found that Phil did not affect the apoptosis of FLSs at the given doses (Fig. 5F).

3.4 The transcripts of inflammatory cytokines and MMP-9 in arthritic FLSs were suppressed by Phil treatment

We demonstrated that Phil could play an anti-inflammatory role in CIA rats. To further confirm this inhibitory effect, we used real-time PCR to detect the mRNA expression of multiple pro-inflammatory factors, including TNFα, IL-1β, IL-6, and IL-8. The results showed that TNFα stimulated the up-regulation of these inflammatory mediators, while Phil treatment significantly decreased their expression. RA FLSs contribute to the degradation of connective tissue by secreting substantial amounts of MMPs, which leads to the breakdown of cartilage. The real-time PCR data from our study showed that Phil had no significant effect on the expression of MMP-2, but effectively inhibited the production of MMP-9 (Fig. 6).

Close

Fig. 6

Phil treatment suppresses the transcripts of inflammatory cytokines and MMP9 in TNFα-induced arthritic FLSs. FLS cells were treated with Phil for 1 h, followed by stimulation with TNFα for 24 h. Real-time PCR was then utilized to assess the mRNA expression of various inflammatory cytokines (IL-1β, IL-6, IL-8, and TNFα) and MMPs (MMP2 and MMP9). Data are presented as means ± SD of three independent experiments (***P < 0.001).

Export to PPT

3.5 Phil treatment blocked TNFα-induced activations of NF-κB and MAPKs

To explore the mechanisms underlying Phil’s inhibition on FLS cells, TNFα-stimulated activation of MAPKs was examined in arthritic FLSs with or without Phil treatment. As shown in Fig. 7A and 7B, the phosphorylation levels of p38, JNK, and ERK were markedly up-regulated in TNFα-induced FLSs. However, all these up-regulations were clearly blocked by Phil treatment (Fig. 7A and 7B). In addition to MAPKs, we also evaluated the activation of NF-κB pathway, which plays a crucial role in the pathogenesis of RA (Li and Makarov, 2006). As shown in Fig. 7C and D, TNFα stimulation caused a significant degradation of IκBα, an inhibitor of NF-κB. however, Phil treatment dose-dependently inhibited this degradation process. In consistency with these data of Western blot, the results of immunofluorescent staining showed that Phil effectively blocked the nuclear translocation of p65 (Fig. 7E), further confirming the inhibition of Phil on the TNFα-induced NF-κB pathway.

Close

Fig. 7

Phil inhibits the activations of MAPKs and NF-κB in TNFα-stimulated arthritic FLS cells. (A and C) FLS cells were treated with Phil at the specified doses for 1 h, followed by stimulation with TNFα for 15 min. Western blot was used to detect the activation of MAPK and NF-κB using the indicated antibodies. (B and D) The relative expressions of p-p38, p-ERK, p-JNK and IκBα were determined by densitometric analysis using Image J. (means ± SD, three independent experiments, **P < 0.01 and ***P < 0.001). (E) FLS cells were treated with 5 μM Phil for 4 h, followed by stimulation with 50 ng·ml⁻¹ TNFα for 30 min. Representative laser confocal images demonstrated that Phil inhibited the nuclear translocation of p65.

Export to PPT