Puerarin Alleviates Experimental Autoimmune Thyroiditis by Regulating Macrophages

Hashimoto’s thyroiditis (HT), also called chronic lymphocytic thyroiditis, was first discovered by Hakaru Hashimoto in 1912. HT is an organ-specific autoimmune disease with genetic susceptibility, mainly causing hypothyroidism, particularly in women (1). HT, as an autoimmune thyroid disease, is characterized by the infiltration of thyroid Ag-reactive immune cells and autoantibodies against thyroid Ags (thyroid peroxidase [TPO], thyroglobulin [TG]) (2). HT is diagnosed by the clinical manifestations of abnormal thyroid function, ultrasound examination showing reduced thyroid parenchymal echo, and increased levels of serum autoantibodies (anti-TG and anti-TPO Abs) (3). The performance of thyroid function can be divided into early and late stages. In the early stage, the release of a large amount of thyroid hormones caused by injury and inflammation of the thyroid leads to thyrotoxicosis (4), whereas hypothyroidism is observed in the late stage, with the increase in thyroid damage (5).

Currently, apart from symptomatic treatment, there are no effective treatments for patients with HT at the late stage. The clinical recommendation is long-term oral administration of l-thyroxine to alleviate symptoms (6). However, the use of l-thyroxine remains controversial mainly because of its side effects in clinical applications (7). Thyroidectomy is the preferred treatment in cases where the goiter of patients with HT is so severe that it affects swallowing and other functions (8). However, traumatic surgery is associated with complications, as well as the continued need to take thyroxine for life after surgery, which degrades the patient’s quality of life (9).

Therefore, in addition to symptomatic treatment and improvement of current symptoms, research on new HT treatment strategies is needed. The limited treatment options for HT are largely due to the lack of understanding of the underlying immune mechanism. Therefore, in addition to symptomatic treatment and improvement of current symptoms, novel HT treatment strategies are urgently required.

Puerarin, an important active isoflavone glycoside extracted from the roots of Pueraria lobata, has been widely used to treat cardiovascular and cerebrovascular diseases (10). In recent years, there have been new findings on the role and mechanism of type 2 diabetes and its complications (11, 12). It has been reported that the traditional Chinese medicine puerarin can improve glucose homeostasis and diabetes by regulating intestinal flora and mucosal immunity (13). Studies have also found that puerarin has an anti-inflammatory effect, improving liver injury and increasing the survival rates of mice with sepsis (14). Therefore, we aimed to validate the effect of puerarin for treating HT in vivo and in vitro to explore new intervention and treatment strategies for HT.

Human thyroid tissue samples from four individuals with HT and four healthy control tissue samples (from thyroid paracancerous normal tissue) were collected from the Department of Pathology, Hospital of Integrated Traditional Chinese and Western Medicine, Nanjing University of Chinese Medicine. In addition, we also obtained normal human peripheral blood in the hospital for subsequent human macrophage experiments. Informed consent was obtained from the subject population for all samples. The research followed guidelines of the Declaration of Helsinki and Tokyo for humans, and was approved by the Ethics Committee of Integrated Traditional Chinese and Western Medicine Hospital (2023SEZ-010).

In this study, we analyzed RNA sequencing data from 20 national samples of 10 patients with HT and 10 healthy controls generated in a previous study from the Gene Expression Omnibus database with accession number GSE153434 (https://www.ncbi.nlm.nih.gov/geo). The R package xCell was applied to estimate the macrophage-related scores and macrophage-related gene signatures (15). The macrophage-related gene signatures are listed in the Supplemental Table I.

Female C57BL/6 mice (specific pathogen-free), aged 6–8 wk, were purchased from Jiangsu Jicui Yaokang Biotechnology. The mice were housed under controlled conditions (12-h light/12-h dark cycle at 22 ± 2°C) and fed with standard food and pure water. All procedures were performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals (Nanjing University) and were approved by the Institutional Animal Care and Use Committee of Nanjing University. The mice were randomly divided into the normal control (NC) group (n = 16), HT group (experimental autoimmune thyroiditis [EAT] model group, n = 20), and HP group (EAT model+puerarin treatment group, n = 20). The mice in the HT and HP groups were s.c. injected with porcine TG (200 μg/mouse) dissolved in CFA (200 μg/mouse) in the neck. Two weeks later, a second immunization was performed, and the mice in the HT and HP groups were injected with porcine TG (200 μg/mouse) dissolved in IFA (200 μg/mouse) into the neck. Simultaneously, mice in the HP group were i.p. injected with 160 mg/kg puerarin (dissolved in PBS) every week, whereas mice in the NC and HT groups were i.p. injected with the same amount of PBS for the initial immunization until the end of the immunization. Puerarin injection was purchased from Shandong Fangming Pharmaceutical Group (national drug approval no. H20033292). All mice were sacrificed 4 wk after the second immunization.

Tissues were fixed in 4% paraformaldehyde solution for H&E staining. Three experimenters blindly evaluated the H&E-stained thyroid sections to determine the mononuclear cell infiltration index score. Mononuclear cell and lymphocytic infiltration were graded as follows: 0, no infiltration; 1, interstitial accumulation of cells between two or three follicles; 2, one or two foci of cells at least the size of one follicle; 3, extensive infiltration, 10–40% of the total area; 4, extensive infiltration, 40–80% of the total area; and 5, extensive infiltration, >80% of the total area. The infiltrating immune cells in the mouse thyroid were evaluated for CD45 (catalog no. 70275, Cell Signaling Technology) expression by immunohistochemistry staining. The thyroid was fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned into 5-μm slices. After Ag retrieval, the slides were blocked with goat serum and incubated with primary Ab overnight at 4°C. An Elite ABC kit and diaminobenzidine substrate were used for immunohistochemistry analysis.

For immunofluorescence analysis, 4% paraformaldehyde-fixed brain tissues were embedded in paraffin and sectioned into 7-μm slices. The slides were blocked with goat serum and incubated with primary anti-CD14 (catalog no. ab221678, Abcam, Waltham, MA) and anti-CD68 (catalog no. ab283654, Abcam, Waltham, MA) Abs overnight at 4°C. Alexa Fluor 594 (catalog no. ab150116, Abcam, Waltham, MA) and Alexa Fluor 488 (catalog no. ab150113, Abcam, Waltham, MA) were used as secondary Abs. After incubation in the dark for 1.5 h, the slides were washed three times with PBS for 5 min each time. The thyroid immunofluorescence images were captured using a confocal microscope (Olympus, Tokyo, Japan).

The HT marker Abs TPO (catalog no. CSB-E08353m, Cusabio, Wuhan, China) and TG (catalog no. CSB-E08241m, Cusabio, Wuhan, China) contained in the eye serum of mice were tested using ELISA kits according to the manufacturer’s instructions. The serum concentrations of IL-10 (catalog no. E-MSEL-M0031, Elabscience, Wuhan, China), IL-17A7 (catalog no. E-MSEL-M0006, Elabscience, Wuhan, China), and CCL2 (catalog no. E-EL-M3001, Elabscience, Wuhan, China) were determined using an ELISA kit according to the manufacturer’s recommendation. RAW264.7 cell supernatants were collected to evaluate the levels of TNF-α (catalog no. E-MSEL-M0002, Elabscience, Wuhan, China), IL-6 (catalog no. E-MSEL-M0001, Elabscience, Wuhan, China), and IL-1β (catalog no. E-MSEL-M0003, Elabscience, Wuhan, China) by ELISA in accordance with the manufacturer’s instructions. The absorbances were measured at 450 nm using an ELISA reader (BioTek Instruments, Baden-Württemberg, Germany).

Total RNA was isolated from the thyroid tissues of mice using TRIzol reagent (Vazyme Biotech, Nanjing, China), before being reverse transcribed to cDNA using the PrimeScript RT reagent kit (Vazyme Biotech, Nanjing, China), and stored at –20°C. Real-time quantitative PCR was performed to detect mRNA levels; cDNA was amplified using SYBR Green Mix (Vazyme Biotech, Nanjing, China) on a ViiA 7 system (Applied Biosystems, Carlsbad, CA). The primer sequences used to amplify the target genes are listed in Table I.

Lymph nodes were collected from the heads and necks of mice, and the cells were squeezed out using glass slides. Subsequently, a 70-μm filter was used to filter the tissue fragments and collect the cell suspension. All collected cells were washed in PBS containing 1% FBS and then adjusted to 1 × 10⁶ cells per 100 µl of PBS. The cells were treated with FITC-labeled anti-mouse F4/80 (catalog no. 123113, BioLegend) and PE-labeled anti-mouse inducible NO synthase (iNOS; catalog no. 696803, BioLegend). All cells were incubated with Abs for 30 min at 4°C in the dark and washed twice in PBS before flow cytometric detection. Analysis was performed using a BD FACSCalibur flow cytometer, and the results of flow cytometry were analyzed using FlowJo software.

The murine macrophage cell line RAW264.7 was purchased from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). The cells were grown in DMEM supplemented with 10% (v/v) FBS and 1% penicillin/streptomycin (Life Technologies, Carlsbad, CA) in humidified incubators (Thermo Fisher Scientific, Waltham, MA) at 37°C under 5% CO₂. The cells were divided into the following four groups: NC group, IFN-γ group, IFN-γ+puerarin treatment group, and puerarin group. The cells were treated with IFN-γ or IFN-γ+puerarin (the concentrations for each are shown in the Fig. 5) for 24 h and lysed for quantitative RT-PCR and Western blot analyses. The cell supernatants were collected to evaluate the concentration of inflammatory markers by ELISA. For the viability test, the cells were seeded at a density of 1 × 10⁵ cells/ml in 96-well plates (five replicates) for 24 h, before challenging with IFN-γ (Sigma-Aldrich, St. Louis, MO) and puerarin for 24 h. Finally, 10 μl of Cell Counting Kit-8 (CCK-8; Beyotime Technology, Beijing, China) solution was added to the cell culture medium in one well of a 96-well plate and incubated for 2 h. The absorbance at 450 nm was measured using a microplate reader (Thermo Fisher Scientific, Waltham, MA).

Thyroid tissue was obtained, subsequently diced, and then subjected to filtration through a strainer and enzymatic digestion using a tissue dissociation kit (Miltenyi Biotec). Cell suspension was added to Percoll centrifugation and the target layer cells were aspirated. CD11b⁺ cells were obtained according to the immunomagnetic bead sorting method (16) using CD11b microbeads kit (Miltenyi Biotec). Cells were inoculated in six-well plates and referred to PBMCs for processing and grouping procedures, followed by cell collection for quantitative real-time PCR (qPCR) experiments.

Human PBMCs were isolated from whole blood using aseptic techniques under aseptic conditions according to the laboratory guidelines (17). The cells were grown in Life Technologies RPMI 1640 culture medium supplemented with 10% (v/v) FBS and 1% penicillin streptomycin (Life Technologies, Carlsbad, CA) in humidified incubators (Thermo Fisher Scientific, USA) at 37°C under 5% CO₂. After the cells were stable, they were resuspended and transferred to a six-well plate to continue culturing. M-CSF (40 ng/ml) was used to stimulate PBMCs for 6 d to induce macrophages. Macrophages were divided into four groups as follows: NC group, IFN-γ group, IFN-γ+puerarin group, and puerarin group. The cells were treated with IFN-γ or IFN-γ+puerarin for 24 h, and lysed for qPCR.

Western blotting was performed to detect the expression of pyroptosis-related proteins. The protein concentrations were determined using a bicinchoninic acid kit (Thermo Fisher Scientific, Waltham, MA) according to the manufacturer’s instructions. The protein (50 µg) from each sample was resolved and then transferred to a polyvinylidene difluoride membrane. The membrane was blocked with 5% BSA for 2 h at room temperature, before washing three times and incubating with IL-1β (catalog no. 2022, Cell Signaling Technology), cleaved caspase-1 (catalog no. 4199, Cell Signaling Technology), caspase-1 (catalog no. 2225, Cell Signaling Technology), p65 (catalog no. 8242S, Cell Signaling Technology), p-p65 (catalog no. 3033S, Cell Signaling Technology), p-STAT3 (catalog no. 9145P, Cell Signaling Technology), STAT3 (catalog no. 9132, Cell Signaling Technology), and GAPDH (catalog no. 110603, Bioword Technology) Abs (all 1:1000) at 4°C overnight. Following incubation, the membrane was incubated with HRP-conjugated goat anti-rabbit Abs (catalog no. AB65151, Bioworld) at a dilution of 1:40,000–1:50,000 for 90 min at room temperature. HRP-conjugated Abs were detected using an ECL Plus blotting reagent and a Quality One documentation system (Bio-Rad Laboratories, Hercules, CA).

All animals were randomly divided into the following four groups: HT group (EAT model group), control liposome group (EAT model+control macrophage scavenger group), clodronate liposome group (EAT model+macrophage scavenger group), and clodronate liposome+puerarin (EAT model+macrophage scavenger+puerarin group). The first cervical s.c. injection of porcine TG (200 μg/mouse) dissolved in CFA (200 μg/mouse) was regarded as the first day. The next day, the clodronate liposome group, clodronate liposome+puerarin group, and control liposome group were injected with clodronate liposome (150 μl/mouse) or its control reagent via the tail vein. Two weeks later, the mice were immunized with IFA (same as the EAT model mentioned above), and the corresponding groups were injected with clodronate liposomes (100 μl/mouse) or their control reagents again the next day. This dose was selected on the basis of the recommendations of the manufacturer. After 2 wk, the mice were sacrificed for harvesting.

Statistical data were analyzed using GraphPad Prism software version 6.0 (GraphPad Software, La Jolla, CA) and SPSS 13.0 software. Two-way ANOVA and multiple comparisons, followed by a Tukey’s test, were used when two factors were involved. Data are expressed as the mean ± SEM. All statistical tests were two-sided, with p < 0.05 considered statistically significant.

According to bioinformatics analysis of the peripheral blood leukocyte transcriptome data in the Gene Expression Omnibus database, the proportion of macrophages in patients with HT was significantly increased, and the expression level of M1-type macrophages was significantly higher than that in the control group (p < 0.001) (Fig. 1A). This indicates that macrophages play a crucial role in the occurrence and development of HT. Further verification in the thyroid tissue of patients with HT using immunofluorescence demonstrated an increase in CD68⁺CD14⁺ macrophages, indicating that the number of proinflammatory M1 macrophages was increased in patients with HT (Fig. 1B). These results of the pathological fluorescence staining of thyroid tissue are consistent with those of the bioinformatics analysis.

We next used the EAT model to construct the phenotype of HT mice, as explained in Materials and Methods (Fig. 2A). From the thyroid pathology, puerarin treatment reduced the degree of lymphocyte infiltration in the thyroid tissues of the HT group; therefore, the infiltration degree score index was significantly reduced in the HP group compared with the HT group (p < 0.05) (Fig. 2B). Moreover, the expression of CD45 was elevated in the thyroid tissues of the HT mice, which was reversed after puerarin treatment (Fig. 2C), suggesting that the thyroid pathology in the HT mice is improved by puerarin treatment. From the ELISA, the expression levels of TG and TPO Abs were significantly higher in the HT group than in the NC group (p < 0.01 and P＜0.05), whereas puerarin treatment significantly reduced the expression levels of autoantibodies in the HP group (p < 0.05).

Previous studies have shown clear evidence that systemic and local inflammation is one of the main mechanisms of the development of HT (1, 18). In this study, we investigated the potential role of puerarin in suppressing inflammation in vivo. Quantitative RT-PCR analysis showed that puerarin significantly inhibited the expression of chemokines and adhesion molecules (CXCL10 and ICAM1) in the thyroid tissues of the HT group (p < 0.05) (Fig. 3A, Table I), suggesting that puerarin effectively inhibited inflammation in thyroid tissues. Puerarin treatment upregulated the transcription levels of the regulatory T cell–related factors Foxp3 and IL-10 (Fig. 3B, 3C), suggesting that puerarin improves local inflammation and immune imbalance in the thyroid tissues of the HT group. Meanwhile, puerarin inhibited the transcription levels of the Th17-type cell-related factor IL-17A and macrophage-related factor CCL2 in the thyroid tissues of the HT group (Fig. 3D, 3E). Moreover, the expressions of IL-10, IL-17A, and CCL2 were significantly increased in the HT group, and puerarin treatment inhibited the expressions of IL-17A and CCL2 and increased the expression of IL-10 (p < 0.01) (Fig. 3F), suggesting that puerarin suppressed systemic inflammation in the HT group. To further investigate macrophages changes in HT, we isolated CD11b⁺ macrophages from thyroid tissue to detect the above genetic changes by quantitative RT-PCR. Interestingly, we found that the inflammatory genes iNOS, IL-6, IL-17A and adhesion molecules ICAM1 were highly expressed in macrophages in the HT group, and the inflammatory levels were reduced after puerarin intervened (Fig. 3G, 3H). The anti-inflammatory factor IL-10 showed increased reactivity in the HT group, and puerarin treatment downregulated IL-10 levels to some extent (Fig. 3I). In addition, we examined the macrophage surface-expressed molecules CD206, CD163, and CD86, which are important in the regulation of immune response and inflammation. The results showed that CD206 and CD86 were significantly upregulated in the HT group, suggesting an immune activation of macrophages and an inflammatory response taking place. CD206 and CD86 were downregulated, and inflammatory levels were suppressed after puerarin treatment (Fig. 3J).

To further explore whether puerarin halted the polarization of M1 macrophages and protected the thyroid from immune injury, we extracted thyroid tissues from mice to analyze macrophage typing by qPCR. The results revealed a low level of iNOS and CD86 in the HP group compared with the HT group, whereas the levels of CD163 and CD206 significantly increased in the HP group compared with the HT group (p < 0.001 and p < 0.01) (Fig. 4A). This suggested that puerarin reduced and increased the proportions of M1 and M2 macrophages, respectively, thereby inhibiting inflammation. The distribution of macrophages in the head and neck lymph nodes of mice was detected by flow cytometry. According to the results, compared with the control group, the proportion of M1-type macrophages (F4/80⁺iNOS⁺) in the HT mice was significantly increased, but was significantly decreased after puerarin treatment (p < 0.05) (Fig. 4B). The immunofluorescence results showed that the expression of F4/80 in the thyroid tissue of HT mice was significantly increased, but was decreased following puerarin treatment (p < 0.01) (Fig. 4C), suggesting that puerarin inhibited the aggregation of macrophages in the thyroid tissue of HT mice.

We next conducted a CCK-8 assay to determine whether puerarin can protect thyroid tissues from inflammatory injury. The results showed that at a puerarin concentration of 20 μM, the cell viability began to recover due to improved resistance to the damaging effect of IFN-γ (Fig. 5A). We next analyzed the effect of puerarin on macrophage polarization. The IFN-γ (20 ng/ml)–induced mRNA expression of M1 marker genes, including iNOS, TNF-α, Il-1β, and IL-6, was reduced by puerarin (Fig. 5B, 5C). In contrast, puerarin increased the expression levels of the anti-inflammatory factor IL-10 in the IFN-γ–induced macrophage injury model (Fig. 5D). We collected RAW264.7 cell supernatants after IFN-γ stimulation in a culture system with or without puerarin to measure the secretion of inflammatory factors by ELISA. The results showed that puerarin significantly reduced the production of IL-6, TNF-α, and IL-1β compared with the IFN-γ stimulation group (p < 0.01, p < 0.05, and p < 0.01) (Fig. 5E). These results indicate that puerarin reduces the secretion of inflammatory factors in vitro.

To further explore the mechanism by which puerarin alleviates inflammatory damage to macrophages, we investigated the pyroptosis signaling pathway in RAW264.7 cells based on recent single-cell sequencing findings in patients with HT. As shown in Fig. 5F, the levels of caspase-1, cleaved caspase-1, and IL-1β in RAW264.7 cells were significantly increased after IFN-γ challenge, all of which were suppressed by puerarin in vitro, as determined by Western blotting (p < 0.01) (Fig. 5F).

In addition, the cell experiments showed that IFN-γ induced the inflammatory response of macrophages, activated p-STAT3, increased the expression of NF-κB, and downregulated p-STAT3 and NF-κB after puerarin intervention, suggesting that puerarin has anti-inflammatory effects. These results demonstrate that puerarin effectively inhibits the pyroptosis pathway in IFN-γ–challenged macrophages (Fig. 5G), confirming that puerarin suppresses inflammation by inhibiting the pyroptosis signaling pathway.

To determine whether puerarin affects M1 versus M2 polarization in human macrophages. We used puerarin to treat M1 human macrophages after IFN-γ stimulation. The results showed that puerarin significantly inhibited the inflammatory genes iNOS and IL-6 (Fig. 5H) and upregulated the anti-inflammatory genes Arg1 and CD206 (Fig. 5I), which were highly consistent with the results of murine-derived macrophage responses.

Puerarin dampens M1 macrophage polarization in vivo and in vitro, suggesting that macrophages play an important role in the development of HT (19). However, it remains unclear whether macrophages are responsible for puerarin-mediated relief of HT; therefore, we next designed macrophage depletion experiments (Fig. 6A). As clodronate liposomes can significantly deplete macrophages in vivo, we compared HT alleviation in mice treated with puerarin or puerarin plus clodronate liposomes. The HT group mice treated with control liposomes presented several HT symptoms, as evidenced by the pathology of thyroid tissues and increased lymphocyte infiltration scores, both of which were alleviated by puerarin (Fig. 6B, 6C). Importantly, the protective effect of puerarin on HT was not stronger in the presence of clodronate liposomes, suggesting that puerarin inhibited HT through a macrophage-dependent mechanism. Additionally, the expression levels of TG and TPO Abs were lower in the clodronate liposomes group and clodronate liposomes+PUE group than in the HT group (Fig. 6D, 6E). However, there was no significant difference in the effect between the clodronate liposomes and clodronate liposomes+PUE groups, further proving that puerarin acts through macrophages.

Chronic lymphocytic thyroiditis (HT) is an organ-specific autoimmune disease whose pathogenesis has not yet been fully elucidated. Through bioinformatics analysis, we found that compared with the control group, the proportion of circulating mononuclear macrophages increased significantly in patients with HT. Further analysis of M1 macrophages showed that the proportion of peripheral M1 macrophages in patients with HT was significantly higher than that in the NC group. To verify the bioinformatics results, we examined the expression of macrophages in histopathological sections of thyroid tissue from patients with HT. The results showed that compared with the NC thyroid tissue, the thyroid tissues of patients with HT exhibited increased infiltration of macrophages and a greater proportion of M1 macrophages.

Consistent with our results, previous studies have revealed the presence of inflammatory macrophages in the thyroid tissues of patients with HT (20, 21). For example, a single-cell sequencing study showed that infiltrating immune cells in the thyroid tissue of HT patients accounted for 65–82% of all cells, mainly T and NK cells, B cells, and plasma cells. However, sequencing results also demonstrated the presence of inflammatory macrophages and dendritic cells expressing high levels of IL-1β in the thyroid, which may be an important cause of thyroid cell destruction in HT patients (21). These inflammatory macrophages may promote the differentiation of various types of immune cells and the destruction of thyroid follicular cells in HT thyroid tissues by secreting a variety of inflammatory factors and chemokines (21). For example, inflammatory macrophages can promote massive activation of autoreactive lymphocytes through Ag presentation and costimulatory signaling (e.g., CD80 and CD86), cytokine secretion, feedback regulation, and cell–cell contact pathways, resulting in destruction of thyroid tissues and increased inflammatory responses (22, 23). In addition, in HT, the metabolism of macrophages is abnormal, and the polarization of macrophages is involved in the process of HT (24). M1 macrophage polarization in HT mice has been reported, and promoting polarization toward M2 macrophages may reverse HT in mice (25). The specific mechanism of the polarization state and function of macrophages in the HT process is incompletely understood, but the regulation of the polarization and function of macrophages may impact the treatment of HT (26, 27). Therefore, future studies should investigate effective drugs that can act directly on macrophages and prevent the development of HT by regulating the function of macrophages.

Puerarin is a derivative of isoflavones isolated from the traditional Chinese medicine Pueraria, which is found in the roots of the legume Pueraria lobata (Willd) Ohwi and Pueraria thunbergiana Benth (28). Puerarin has antipyretic and sedative effects, as well as the ability to increase coronary blood flow (29). Clinically, puerarin is used to treat coronary heart disease, angina pectoris, hypertension, and other diseases (30). We previously demonstrated that puerarin effectively improved high-fat diet-induced type 2 diabetes and effectively prevented the death of endotoxin-induced sepsis mice (31). We also found that puerarin can modulate macrophage polarization and inflammatory signaling pathways, indicating that it can be used in metabolism- and immunity-related diseases (32). In addition, puerarin can also regulate the immunological activity of lymphocytes and inhibit lymphocyte-mediated inflammatory responses by modulating cytokine production, regulating the expression of cell-surface receptors or modulating signaling pathways, and by its anti-inflammatory effects (33–35). On the basis of the above, we hypothesized that puerarin could serve as an effective treatment for HT. Therefore, in this study, we used the classic EAT model to simulate HT, although the animal model could not completely simulate the dynamic changes of HT in clinical practice, and was less complex than clinical HT (36). However, the animal model exhibited an increase in autoantibodies and pathological changes in thyroid tissue (infiltration of a large number of inflammatory cells), providing an appropriate experimental model for our subsequent studies. Consistent with our hypothesis, puerarin effectively inhibited the level of autoantibodies and reduced the infiltration of inflammatory cells in the thyroid tissues of HT mice.

Cell experiments showed that IFN-γ induced the inflammatory response of macrophages, activated p-STAT3, and increased the expression of NF-κB. NF-κB and p-STAT3 were downregulated after puerarin interference, suggesting that puerarin directly inhibited the inflammatory activation of macrophages. It has been reported that the presence of inflammatory macrophages expressing high levels of IL-1β in the thyroid contributes to thyrocyte destruction in patients with HT (21, 37). Our results showed that high levels of IL-1β existed in both in vivo HT model mice and in vitro IFN-γ–stimulated cell models. Simultaneously, based on the single-cell sequencing results of patients with HT, the pyroptosis signaling pathway in RAW264.7 cells was studied. The results indicated that cleaved caspase-1 increased with the increase in IL-1β expression. However, puerarin treatment reduced the expression of pyroptosis protein cleaved caspase-1, inhibited the pyroptosis signaling pathway activated during thyroiditis, and alleviated the damage to HT to a certain extent.

Although we demonstrated that puerarin directly acts on macrophages, inhibiting the release of inflammatory factors and altering the polarization state, the specific target or receptor of puerarin on macrophages remains unclear. Previous studies have reported several possible mechanisms. 1) Anti-inflammatory effect: the main target of puerarin in macrophages is the TLR4/NF-κB and NLRP3/IL-1β signaling pathway (38, 39), which regulates the expression of proinflammatory cytokines and chemokines (40). 2) Antioxidant effect: puerarin can increase the activity of antioxidant enzymes, such as SOD and CAT, in macrophages (41, 42), reduce the production of reactive oxygen species, and protect cells from oxidative stress damage (43). 3) Immunoregulatory effect: puerarin can regulate the immune function of macrophages by activating immune receptors, phagocytic activity, and Ag presentation function (33, 44). Puerarin can regulate the function of immune cells and promote the maturation and activation of APCs (e.g., dendritic cells) to enhance their ability to recognize and present self-antigens, and, at the same time, the anti-inflammatory effect of puerarin helps to maintain the normal function of the APCs and reduces inflammation’s interference with their ability to present self-antigens (44–47). For instance, the mu-opioid receptor is an important G protein–coupled receptor involved in many physiological and pathological processes and is widely studied in the fields of pain, depression, and drug addiction (48). Puerarin activates the mu-opioid receptor in macrophages, thereby promoting immune regulatory function (49). 4) Antibacterial effect: puerarin can inhibit the macrophage-mediated phagocytosis and killing of certain bacteria, thereby enhancing the body’s immune function (19, 50).

Considering that puerarin can inhibit the polarization of M1 macrophages both in vivo and in vitro, we aimed to further verify the effective mechanism of puerarin. We then exhausted macrophages with clodronate liposomes and compared the alleviating effects of puerarin or puerarin plus clodronate liposomes on HT in mice. Interestingly, the protective effect of puerarin on HT was weak in the presence of clodronate liposomes, suggesting that puerarin inhibits HT through a macrophage-dependent mechanism. In addition to targeting macrophages, puerarin has certain effective effects on other host cells; thus, in terms of its mechanism in HT, it cannot be ruled out that puerarin synergically mediates functional changes of other cells (19). At the very least, we confirmed that puerarin alleviates HT progression primarily by regulating the polarization and function of macrophages.

In summary, our study confirmed that puerarin alleviated the disease phenotype of the HT mice and inhibited systemic and local inflammatory responses, especially the proportion of macrophages in the HT group. In vitro experiments also confirmed that puerarin can directly affect IFN-γ–stimulated macrophages, inducing their polarization to M2, and inhibiting the pyroptotic signaling pathway of macrophages. Finally, we used clodronate liposomes, which can significantly deplete macrophages in vivo, to verify that the target of puerarin treatment is partially dependent on macrophages. Our findings confirm the role of puerarin in organ-specific autoimmune diseases and provide new insights into treatment strategies for HT.

The authors have no financial conflicts of interest.

We thank Prof. Qian Gao from the School of Medicine of Nanjing University for guidance, ideas, and research support for this study.