Rheumatoid arthritis (RA) is an autoimmune disease that attacks the body’s joints and is characterized by swelling, pain, synovial inflammation, and the destruction of the joint structure. Its pathology is characterized by chronic and multiple localized infiltrations of inflammatory cells in the joints and the proliferation of synovial tissue, eventually leading to cartilage and bone destruction. The exact pathogenesis of RA is unclear; however, studies have shown that immune cell imbalance is critical in its development.

Macrophages can polarize into two phenotypes with diametrically opposing functions in different microenvironments: M1 and M2 type. These macrophages are the main inflammatory cells in the synovium and are crucial in the pathology of synovitis [1]. Synovial macrophages and their inflammatory factors are decisive in the inflammatory response and joint destruction in osteoarthritis [2]. Removing synovial pro-inflammatory macrophages to reestablish macrophage homeostasis has been shown to improve RA and other autoimmune diseases [3]. In RA, there is an imbalance in the M1/M2 ratio, with tissue and organ damage caused mainly by an increase in M1-type macrophages. M1 macrophages participate in inflammatory response by producing interleukin 6 (IL-6), tumor necrosis factor α (TNF-α) and inducible nitric oxide synthase (iNOS), whereas M2 macrophages inhibit inflammatory response and promote tissue damage repair by producing interleukin 10 (IL-10) and arginase 1 (Arg-1). Previous studies have demonstrated that macrophage polarization is essential in RA development. Therefore, exploring the mechanisms underlying macrophage polarization may provide new therapeutic targets and strategies for treating inflammatory joint diseases.

According to the literature, mitogen-activated protein kinase (MAPK) phosphatase 1 (MKP-1) deficiency exacerbates disease progression in mice with RA [4]. Furthermore, MKP-1 is critical in suppressing inflammation by promoting macrophage polarization towards the M2 phenotype [5]. Moreover, in macrophages, MKP-1 is thought to be a key negative regulator of macrophage MAPK signaling in response to inflammatory stimuli, dephosphorylating extracellular signal-regulated kinase (ERK), p38, and c-Jun amino-terminal kinase (JNK).

Retinoic acid is a biologically active metabolite or derivative of vitamin A. It exists as various isomers, including all-trans retinoic acid. Retinoic acid can regulate cell growth and differentiation and induce apoptosis, which is widely used in treating dermatological diseases and some tumors. Spleen M1 macrophages in patients with immune thrombocytopenia are abnormally polarized and retinoic acid may play a therapeutic role by restoring the balance of macrophages to M2 [6]. Currently, it has been shown that retinoic acid can turn lipopolysaccharide (LPS)-induced macrophages away from M1 polarization by negatively regulating the nuclear factor kappa light chain enhancer of activated B cells (NF-κB)/microRNA-21 signaling pathway [7]. Retinoic acid also attenuated the transmissible gastroenteritis coronavirus infection-induced apoptosis in intestinal porcine enterocyte cells by inhibiting the p38/MAPK signaling pathway [8]. Retinoic acid has also been shown to slow the progression of RA by inhibiting the migration and invasion of fibroblast-like synovial cells [9]. However, the specific mechanisms underlying the anti-arthritic influences of retinoic acid, especially on macrophages, is still unclear.

Therefore, the objective of this research was to explore the anti-arthritic effects of retinoic acid in mice with adjuvant arthritis (AA) and to elucidate the potential mechanisms it promotes. Our results show that retinoic acid attenuates inflammatory immune response and suppresses joint destruction. This study provides new perspectives into the immunomodulatory mechanisms of retinoic acid; this may regulate macrophage polarization through the MKP-1/MAPK pathway, thereby attenuating inflammation.

Drugs and reagents

Retinoic acid (purity ≥ 98%) was purchased from Shanghai Macklin Biochemical Co. Complete Freund's adjuvant and LPS were obtained from Sigma. Methotrexate (MTX) was purchased from Tonghua Maoxiang Pharmaceutical Co Ltd. The transfection kit was purchased from RiboBio Co. Ltd. Flow cytometry anti-iNOS (PE), anti-CD206 (APC), Fixable Viability Dye eFluor 780 and Fix/Perm buffer were purchased from eBioscience. The ELISA kits for TNF-α, IL-6 and IL-10 were manufactured by Mlbio. Primary antibodies included p-JNK, JNK, p-P38, P38, p-ERK, ERK were purchased from Affinity Biosciences. Antibodies against iNOS, Arg-1, cyclooxygenase-2 (COX-2), peroxisome proliferator-activated receptor gamma (PPAR-γ), MKP-1 and β-actin were purchased from Santa Cruz.

Animals and experimental design

Male Balb/c mice were purchased by the Animal Experimentation Center of Yanbian University. The mice were kept at appropriate temperature (22°C ± 2°C) and humidity with food and water provided. This study was approved by the Laboratory Animal Ethics Committee of Yanbian University (Approval No. YD202309110021).

The AA model was established by intradermal injection of 0.05 ml complete freund’s adjuvant (CFA) into the left hind paw [10]. In contrast, mice in the control group were injected with the same dose of saline. There were 24 mice weighing 18–22 g. Animals were randomly divided into 4 groups (n = 6) in each group, including control group, AA model group, MTX group (0.5 mg/kg once every 3 days) and retinoic acid group (15 mg/kg/d). The therapeutic drugs were administered intragastrically for 15 days. The control and model groups were given saline orally during the treatment period.

Cell culture

The collected peritoneal cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) containing 10% fetal bovine serum (FBS). Non-adherent cells were removed after 2 h incubation at 37°C with 5% CO₂ and adherent cells were observed under a microscope. Then the next experiment was carried out.

RAW264.7 macrophages were purchased from the American Type Culture Collection. RAW264.7 was cultured in DMEM (containing 10% FBS and 1% antibiotics) at 37°C in a humidified incubator containing 5% CO₂. Induced RAW264.7 cells were treated with LPS (1 μg/ml) for 12 h. Then cells were incubated with RA (1 μM) for 24 h.

Assessment of paw edema and arthritis index (AI) and thermal nociceptive sensitization test

The thickness of the left hind paw swelling in a mouse was measured using calipers before and after CFA immunization [11]. Recordings were then made every 7 days until the experiment ended. The mice were immobilized and the diameter of the paw thickness was measured by using digital calipers.

The AI was assessed by other independent observers. The true nature of joint inflammation in mice was observed periodically every 7 days after CFA induction until sacrificed. The evaluation criteria for arthritis [12] were as follows: 0 represented no swelling or edema, 1 represented mild edema and limited erythema, 2 represented mild edema and erythema from the ankle to the tarsus, 3 represented moderate edema and erythema from the ankle to the tarsus, and 4 represented edema and erythema from the ankle to the entire leg. The severity of all four limbs (up to a maximum of 16 for individual mice) was summed to give an arthritis score for every mouse.

Thermal nociceptive sensitization was assessed using the Pathway-Cheops Thermal Pain Stimulator, according to the methodology of the literature [13]. Paw withdrawal latency (PWL) was used to assess pain in the mice. PWL was measured by how long it took the mice to withdraw their paw. PWL was recorded prior to CFA injection as baseline latency and at seven-day intervals throughout the experiment as final latency.

Histopathologic features

The ankle tissues of mice were fixed in 10% formalin and decalcified in 10% EDTA. Finally, the treated tissues were embedded in paraffin. Serial paraffin sections were analyzed histopathologically with hematoxylin and eosin as well as Saffron O and solid green staining and examined using a light microscope.

After drying, dewaxing, and hydration, paraffin sections were heated by hyperthermia for antigen repair and incubated with endogenous peroxidase blocker at room temperature, next followed by incubation with primary antibody MMP13 (1:500) at 4°C overnight. The sections were then incubated with the appropriate secondary antibodies for 20 min at appropriate temperature. Incubate slides with the DAB kit for 10 min according to the manufacturer's instructions. Images were observed and captured using microscope.

Splenic index and measurement of inflammatory factors by ELISA

Mice were weighed and then sacrificed at the end of drug administration. Spleens were collected from each group. Spleen index was calculated using the following formula: splenic index = wet weight of spleen (mg) / body weight of mouse (g) [14].

Blood samples were separately collected in different tube at the end of treatment. Concentrations of TNF-α, IL-6 and IL-10 were measured using ELISA kits according to the manufacturer's instructions.

Immunofluorescence and macrophage migration assay

Macrophages were collected by injecting pre-cooled sterile phosphate-buffered saline into the peritoneal cavity. The collected groups of cells were resuspended and then placed at 1 × 10⁵/well in six-well plates placed on coverslips. Macrophages from each group were separately fixed, permeabilized and closed in 1% bovine serum albumin. These samples were then incubated overnight at 4°C with primary antibodies against iNOS and Arg-1. The next day all samples were incubated by fluorescence-conjugated secondary antibodies for 1 h at room temperature. DAPI was used to stain and locate the nucleus. Immunofluorescence images of cells were captured by fluorescence microscope and analyzed semi-quantitatively using Image J analysis software.

Cell migration assays were performed using Transwell chambers (Corning Incorporated). Briefly, complete medium was added in the lower chamber and serum-free medium and cells were in the smaller chamber. The operation was complete and placed in the incubator. Cells with the ability to migrate will cross the chambers. After 24 h, cells migrating to the lower surface were fixed with 4% paraformaldehyde for 15 min. Furthermore, cells were stained with Giemsa stain. In the end, cells were observed under an inverted microscope.

Flow cytometry

RAW264.7 Macrophage suspension (1 × 10⁶/ml) was stained with CD16/32 antibody for 30 min at 4°C in the dark. For intracellular staining, cells were further permeabilized using fixation/membrane-breaking buffer and then stained with iNOS monoclonal antibody (PE) and CD206 monoclonal antibody (APC). Samples was then recorded and analyzed using the CytExpert software.

Cell transfection

The small interfering RNA (siRNA) against MKP-1 (Dusp1) was designed and synthesized by RiboBio. The target sequence was GCATCACCGCCTTGATCAA. RAW264.7 cells were inoculated into six-well cell culture plates at 2 × 10⁵/ml, and the cells were induced to adhere to the wall using DMEM medium containing 10% FBS, and transfection was performed 24 h later. SiRNA-Dusp1 and siRNA-NC were transfected according to the instructions of the riboFECTTMCP transfection reagent.

Western blot

Harvested RAW264.7 macrophage samples were passed through RIPA buffer to prepare protein lysates. For protein blotting, protein samples were separated on a prepared gel. After electrophoresis, the protein was transferred to a PVDF membrane by transmembrane, then blocked with skim milk for 1.5 h, and finally incubated with primary antibodies overnight at 4°C. The blot was then incubated with HRP-conjugated secondary antibody (Boster Biotechnology), detected and imaged with x-ray film.

Statistical analysis

We performed statistical analyses using one-way ANOVA. And two-way ANOVA and Tukey’s post-hoc test was applied. Data in graphs are presented as mean ± SD and were analyzed using GraphPad Prism 8 software. p-values < 0.05 indicate that differences are statistically significant.

Effect of retinoic acid on the overall assessment of mice with arthritis

Mice with AA developed paw swelling, nodules, and redness after a CFA induction. In the group treated with retinoic acid and MTX on day 30, paw thickness and swelling were reduced (Fig. 1B, D). The overall arthritis score in mice with AA was significantly reduced in the retinoic acid (15 mg/kg) and MTX (0.5 mg/kg) groups (Fig. 1C). Immune dysfunction supports RA development. The splenic index of mice with AA was considerably higher than that of normal mice after immunization, as we observed. The splenic index was significantly reduced in the retinoic acid (15 mg/kg) and MTX (0.5 mg/kg) treatment groups compared with the model group (Fig. 1F). We examined PWL in all groups 1 day before modeling and on days 7, 14, 21, and 28 after modeling. The findings demonstrated that compared to the retinoid and MTX-treated groups, the PWL in the model group was significantly lower. These results indicated that retinoic acid alleviated the arthritis-induced pain (Fig. 1E).

Figure 1. Retinoic acid alleviates arthritis. (A) Anti-arthritis treatment schedule. (B) Left hind paw of mice in each group on day 30. (C, D) Overall assessment of arthritis and foot and metatarsal swelling were performed in each group from day 0 to day 28 and assessed at 7-day intervals. (E) Paw withdrawal latency was assessed in the different groups. (F) Spleens were harvested, and splenic indices were determined on day 30 after immunization. Statistical analysis was performed by using two-way ANOVA and Tukey’s post-hoc test. Data are expressed as mean ± SD (n = 6). CFA, complete freund’s adjuvant; MTX, methotrexate; RA, rheumatoid arthritis. ^##p < 0.01 and ^###p < 0.001 vs. control group; *p < 0.05, **p < 0.01 vs. model group.

Retinoic acid improves histopathology in arthritic mice

Histopathological examination of ankle joints in mice with AA revealed synovial tissue proliferation, inflammatory cell infiltration, and cartilage erosion. A scoring system [15] used to evaluate the pathological variations in the ankle joints showed that the retinoic acid (15 mg/kg) and MTX (0.5 mg/kg) treatment groups significantly reduced arthritic symptoms (Fig. 2A, C). As shown in Fig. 2A, Safranin-O staining indicated that cartilage erosion occurred in the AA model group and was ameliorated by retinoic acid and MTX treatment. Matrix metallopeptidase 13 (MMP13) is closely associated with arthritis. Immunohistochemistry results showed a significant increase in MMP13 expression compared to the AA model group and a substantial decrease in MMP13 expression in the retinoic acid and MTX treatment groups (Fig. 2B, D).

Figure 2. Retinoic acid ameliorates histopathology in mice with arthritis. (A) Histopathologic changes were evaluated using H&E and Safranin-O staining (10×, scale bar = 200 μm). a (synovial proliferation), b (cartilage erosion), c (pannus formation), d (inflammatory cell infiltration). The red arrow indicates that the joint space has narrowed. (B) Representative micrographs of matrix metallopeptidase 13 (MMP13) protein expression in IHC (20×, scale bar = 100 μm; 40×, scale bar = 50 μm). (C) The score of histological changes of ankle joints in different groups were assessed. (D) MMP13-positive areas of ankle cartilage. Data were analyzed using one-way ANOVA and Tukey’s post-hoc test. Data are expressed as mean ± SD. Results are represented by at least three independent experiments. MTX, methotrexate; RA, rheumatoid arthritis. ^#p < 0.05, ^###p < 0.001 vs. control group; *p < 0.05 and **p < 0.01 vs. model group.

Effect of retinoic acid on the expression of pro- and anti-inflammatory serum cytokines in AA mice

We used enzyme-linked immunosorbent assay kits to determine the concentrations of the pro-inflammatory cytokines IL-6 and TNF-α and the anti-inflammatory cytokine IL-10 in mouse serum. The serum levels of IL-6 and TNF-α were significantly increased, whereas the IL-10 level was lower in the AA model group than in the control group. After treatment with retinoic acid and MTX, the production of IL-6 and TNF-α was inhibited and that of IL-10 was increased (Fig. 3).

Figure 3. Retinoic acid reduces the secretion of pro-inflammatory factors. (A–C) Tumor necrosis factor α (TNF-α), interleukin 6 (IL-6), and interleukin 10 (IL-10) levels were measured. Data were analyzed using one-way ANOVA and Tukey’s post-hoc test. Data are expressed as mean ± SD (n = 6). MTX, methotrexate; RA, rheumatoid arthritis. ^##p < 0.01 and ^###p < 0.001 vs. control group; **p < 0.01 vs. model group.

Retinoic acid modulates macrophage polarization in AA mice

An imbalance in the macrophage M1/M2 ratio has been proposed as associated with RA. To investigate whether the anti-inflammatory effect of retinoic acid was related to the inhibition of M1 polarization, we examined the expression of M1 macrophage marker (iNOS) and M2 macrophage marker (Arg-1) in peritoneal macrophages of each group. iNOS expression was significantly higher and Arg-1 expression was decreased in the AA model group compared to the control group. In the retinoic acid (15 mg/kg) and MTX (0.5 mg/kg) treatment groups (Fig. 4A, D), iNOS expression levels were significantly reduced, and Arg-1 expression was upregulated (Fig. 4B, E). We also observed macrophage migration in each group, which increased in the AA group and decreased in the retinoic acid and MTX treatment groups (Fig. 4C, F). Our results suggest that retinoic acid exerts anti-arthritic effects by modulating macrophage polarization.

Figure 4. Retinoic acid affects macrophage polarization and function in mice with adjuvant arthritis. (A) Quantification of inducible nitric oxide synthase (iNOS) expression in peritoneal macrophages by immunofluorescence (100× magnification). (B) Quantification of arginase 1 (Arg-1) expression in peritoneal macrophages using immunofluorescence (100× magnification). (C, F) Representative images showing macrophage migration in various groups (200× magnification). (D) Quantitative analysis of iNOS levels. (E) Quantitative analysis of Arg-1 levels. Statistical analysis were performed by one-way ANOVA and Tukey’s post-hoc test. Data are expressed as mean ± SD. Results are represented by at least three independent experiments. MTX, methotrexate; RA, rheumatoid arthritis. ^###p < 0.001 vs. control group; *p < 0.05 and **p < 0.01 vs. model group.

Effect of retinoic acid on LPS-induced RAW264.7 macrophages

We created the model of pro-inflammatory macrophages through the LPS stimulation of RAW264.7 cells to further validate whether retinoic acid could inhibit the polarization of M1 macrophages. To further validate our results, we examined the value of iNOS⁺ cells and CD206⁺ cells in several groups by flow cytometry. The proportion of iNOS⁺ macrophages was higher in LPS-stimulated macrophages than in controls. This effect was reversed upon treatment with retinoic acid. In contrast, in all cases, the proportion of CD206⁺ macrophages in the LPS group was lower than that in the control group. Retinoic acid treatment increased the proportion of M2 macrophages (Fig. 5A–E). Protein blot analysis showed (Fig. 5F) that LPS significantly upregulated the expression of iNOS and COX-2 while inhibiting Arg1 and PPAR-γ (Fig. 5G). As expected, retinoic acid significantly upregulated Arg1 levels and inhibited iNOS levels. Consequently, our in vivo and in vitro experiments demonstrate that retinoic acid attenuates the inflammatory response in an AA mouse model by modulating macrophage polarization.

Figure 5. Retinoic acid inhibited lipopolysaccharide (LPS)-induced M1 polarization in RAW264 .7 macrophages. (A) Removal of dead cells and debris. (B) The gate strategy of iNOS+ macrophages (M1) and CD206+ macrophages (M2). (C–E) The proportion of M1 and M2 macrophages in each group was analyzed by flow cytometry. (F) Protein levels of inducible nitric oxide synthase (iNOS), arginase 1 (Arg-1), cyclooxygenase-2 (COX-2), and peroxisome proliferator-activated receptor gamma (PPAR-γ) were analyzed by Western blotting. (G) Quantitative analysis of iNOS, Arg-1, COX-2, and PPAR-γ protein levels. Statistical analysis were performed by using one-way ANOVA and Tukey’s post-hoc test. Data are expressed as mean ± SD. Results are represented by at least three independent experiments. RA, rheumatoid arthritis. ^##p < 0.01 and ^###p < 0.001 vs. control; *p < 0.05, **p < 0.01 and ***p < 0.001 vs. model group.

Retinoic acid regulates RAW264.7 macrophage polarization in connection with MKP-1 activation

Based on these experimental results, we hypothesized that retinoic acid regulates macrophage polarization dependent on MKP-1 activation. Therefore, the study knocked down MKP-1 using siRNA cellular transfection and constructed the RAW264.7 cell line with low MKP-1 expression. We also investigated whether inhibiting MAPK signaling by retinoic acid depended on MKP-1. Protein blotting analysis showed (Fig. 6) that retinoic acid significantly inhibited the phosphorylation of ERK, JNK, as well as p38 in LPS-induced macrophages; however, the above effect was not observed in MKP-1 knockdown cells. Consistent with our hypothesis, MKP-1 deficiency reversed the effects of retinoic acid in LPS-stimulated macrophages. These results suggest that retinoic acid inhibits the MAPK pathway through MKP-1.

Figure 6. Retinoic acid inhibits the MAPK signaling by upregulating MKP-1. (A) Detection of protein levels of mitogen-activated protein kinase (MAPK) phosphatase 1 (MKP-1), MAPK pathway in macrophages using Western blotting. (B–E) Quantitative analysis of MKP-1, p-JNK/JNK, p-ERK/ERK, p-P38/ P38 expression levels. Statistical analysis were performed by using one-way ANOVA and Tukey’s post-hoc test. Data are expressed as mean ± SD. Results are represented by at least three independent experiments. JNK, c-Jun amino-terminal kinase; ERK, extracellular signal-regulated kinase; RA, rheumatoid arthritis. ^###p < 0.001 vs. control; **p < 0.01 vs. model group.

Studies have shown that an M1/M2 ratio imbalance is involved in RA pathogenesis [16]. Retinoic acid is crucial in controlling inflammatory diseases, and its role in regulating macrophage polarization in arthritis has been less reported; therefore, we investigated the mechanism of its anti-arthritic action. MKP-1, an enzyme that dephosphorylates MAPKs, inactivating signaling cascades that pass through these pathways, strongly regulates macrophage polarization and is involved in various physiological and pathological processes [17,18]. This study demonstrated that the therapeutic effects of retinoic acid are relevant to the regulation of macrophage polarization by MKP-1. In addition, retinoic acid restores macrophage homeostasis by upregulating MKP-1 to promote the dephosphorylation of p-JNK, p-ERK, and p-P38.

The vitamin A analogs retinoic acid and all-trans retinoic acid have been shown to have beneficial therapeutic and preventive effects in various inflammatory diseases [19]. Previous studies have shown that retinoic acid mediates the differentiation of Treg cells through RARα [20]. Treg, which is produced in the presence of retinoic acid, inhibits collagen-induced arthritis [21]. All-trans retinoic acid enhanced the regulatory T cell responses, inhibited T helper 17 (Th17) production and function, inhibited LPS-induced NF-κB activation, and attenuated dextran sodium sulfate-induced experimental colitis. Retinoic acid also exerts anti-arthritic effects by inducing the expansion of regulatory T cells and inhibiting Th17 cells to ameliorate the clinical course of collagen-induced arthritis in mice [22]. Based on these studies, we evaluated the therapeutic effects of retinoic acid in AA mice. Similarly, our current study showed that retinoic acid relieves paw oedema and reduces arthritis scores and spleen index in mice with AA. In addition, the development of RA is accompanied by pain as a result of inflammation. Based on the repeated administration of mazindol, which reduced spontaneous pain-like behavior in mice with CFA-induced arthritis [23], we investigated the effect of retinoic acid on nociceptive hypersensitivity in AA mice. This study demonstrated that retinoic acid relieves the painful symptoms of RA.

Bioinformatic analysis has established that M1-type macrophages play a pro-inflammatory role in the synovial environment of RA [24]. There is an imbalance in the macrophage subpopulations in the synovial fluid of patients with RA, and the M1/M2 ratio is significantly higher than that in patients with osteoarthritis [25]. In addition, it has been reported that Kinsenoside reduces osteoarthritis by repolarizing macrophages by inhibiting the NF-κB/MAPK signaling pathway [26]. Therefore, maintaining the M1/M2 balance can effectively inhibit the joint inflammatory response. Studies have shown that retinoic acid significantly downregulated the expression of iNOS and upregulated the expression of Arg-1 in peritoneal macrophages of AA mice. This suggests that retinoic acid reduces arthritis symptoms by inhibiting macrophage M1 polarization, promoting macrophage polarization towards M2 and modulating the balance between M1/M2. The above data suggest that inhibition of M1 macrophage polarization and promotion of M2 macrophage polarization may be a new strategy for the treatment of RA. The dynamic balance of the macrophage M1/M2 ratio is crucial in maintaining the homeostasis of the internal environment. We established an in vitro inflammatory model using LPS-induced RAW264.7 macrophages to explore the specific retinoic acid-induced macrophage polarization mechanism. Retinoic acid was found to dose-dependently downregulate the expression of iNOS and upregulate the expression of mannose receptor (CD206) and Arg-1 in Pg-LPS-induced macrophages [27]. COX-2 was significantly expressed in LPS-stimulated macrophages and is involved in inflammatory response [28,29]. PPAR-γ is critical in regulating metabolism, controlling inflammation, suppressing tumors, and modulating immune processes. Its expression was correlated with M2 macrophages [30]. Similarly, our study showed that retinoic acid inhibited the expression of iNOS and COX-2 in M1 macrophages and promoted the expression of Arg-1 and PPAR-γ in M2 macrophages. Our in vivo and in vitro findings revealed that retinoic acid might be effective in restoring the balance between M1 and M2 macrophages to regulate inflammation and immune responses.

MKP-1 is a nuclear-localized bispecific phosphatase that dephosphorylates MAPK-ERK, p38, and JNK. MKP-1 is also an important regulator of macrophage behavior and controls inflammation as a molecular tool to regulate the innate immune response [31]. Notably, several studies have demonstrated that MKP-1 plays an anti-inflammatory role in various disease models [32]. MKP-1 activation regulates macrophage polarization, which is involved in muscle repair processes [33]. Previously, dexamethasone was shown to upregulate MKP-1 and dephosphorylate P38 to antagonize the anti-inflammatory effects of microsomal prostaglandin E synthase-1 in mice with RA [34]. In addition, activating MKP-1 subsequently inhibits the phosphorylation of P38, the P38 MAPK involved in immune regulation and inflammatory responses, such as RA [35] and ulcerative colitis [36]. Rapamycin has been shown to negatively regulate macrophage activation by limiting the NLRP3 inflammatory vesicle-p38 MAPK-NF-κB pathway in an autophagy and p62/SQSTM1-dependent feedback loop [37]. Recently, it has been shown that inhibiting the p38 MAPK pathway can attenuate inflammation of joints by suppressing the release of pro-inflammatory macrophage-associated cytokines [38]. Therefore, after retinoic acid treatment, we investigated changes in MKP-1, p-JNK, p-ERK, and p-P38 levels. Our results suggested that the regulatory effect of retinoic acid on macrophage homeostasis may depend on the MKP-1/MAPK pathway.

Retinoic acid is crucial in mitigating retinal degeneration by attenuating photoreceptor loss following blue light exposure through the MKP-1/JNK pathway [39]. To determine whether the influence of retinoic acid on macrophage polarization is associated with MKP-1, we transfected RAW264.7 cells with MKP-1 siRNA. We found that the role of retinoic acid in macrophage polarization and function were not observed in the MKP-1 siRNA group. Notably, the inhibitory effect of retinoic acid on the P38/MAPK pathway was attenuated after MKP-1 knockdown in RAW264.7 cells. This suggests that retinoic acid modulates macrophage polarization by inhibiting MAPK activation in a MKP-1-dependent manner.

In conclusion, our study provides evidence that the retinoic acid-induced alleviation of arthritis symptoms in mice with AA depends on restoring M1/M2 polarization and inhibiting the MAPK pathway. Notably, MKP-1 activation is a critical target of the immunomodulatory function of retinoic acid.

This study was supported by the National Natural Science Foundation of China (82360479), the Natural Science Research Foundation of Jilin Province for Sciences and Technology (YDZJ202301ZYTS173) and the Project of Education Department of the Jilin province of China (JKH20210583K).

None.

The authors declare no conflicts of interest.