B7-H3 Blockade Decreases Macrophage Inammatory Response and Alleviates Clinical Symptoms of Arthritis

Background: The costimulatory molecule B7-H3 is an immunoregulatory protein that plays an important role in the course of autoimmune diseases. It is also involved in the regulation of inammatory responses in the body. Studies have shown that B7-H3 is highly expressed in peripheral blood monocytes of rheumatoid arthritis (RA) patients and its expression is closely related to the clinical parameters of the disease. In this study we aimed to determine what the implication of high B7-H3 expression for the disease severity in human RA and mouse model of arthritis is, and whether targeting this molecule could constitute a new therapeutic approach. Methods: We combined histopathological scoring of the joint cavity of collagen-induced arthritis (CIA) mice with immunouorescent studies of B7-H3 expression on macrophages. We assessed the function of B7-H3 in relation with the pro-inammatory role of macrophage through small RNA interference. Then we studied the molecular pathways liking B7-H3 with the pro-inammatory activity of macrophages by multiplex ow cytometry, western blot and quantitative real-time PCR. The therapeutic benet of anti-B7-H3 blocking was probed in mouse with CIA. Results: Histopathological and histological examination showed a positive correlation between expression of B7-H3 on macrophages and disease activity score, degree of destruction of the joint cavity, and pannus formation. It was also correlated with increased expression of the pro-inammatory cytokine TNF-α in the joint cavity tissue and TNF-α level in peripheral serum. Knocking down B7-H3 expression through stable expression of small interfering RNA in human and mouse mononuclear phagocytic cell lines weakened the inammatory response of the cells. Further molecular analyses indicated that the regulation of the inammatory response through B7-H3 involved NF-κB signaling. Treatment of CIA mice with anti-B7-H3 reduced the clinical manifestation of arthritis and downregulated the expression of pro-inammatory cytokines. Conclusions: We conrmed increased expression of B7-H3 in arthritis and established a positive correlation between disease severity


Background
Rheumatoid arthritis (RA) is an autoimmune disease with symmetric polyarthritis as its main clinical manifestation. It is the most common autoimmune disease, and it is characterized by massive in ammatory cell in ltration, abnormal hyperplasia of synovial tissue, and progressive destruction of articular cartilage and bone [1] . Its clinical characteristics include in ammation of the synovium of the hands, feet, wrists, knees, buttocks, and other joints. The long-lasting and repeated attacks of in ammation lead to the destruction of cartilage and bone in the joints, dysfunction of joints, and even physical disability. In addition, it can be accompanied by vascular in ammation that affects all organs of the body. Seventy percent of RA patients have different degrees of disability within 3 years, and therefore RA is one of the main causes of disability and labor loss [2] . RA seriously reduces human health and quality of life. However, its pathogenesis has not been fully elucidated, and it still lacks speci c treatment [3] .
B7-H3 (also known as CD276), which has been identi ed as a tumor-associated antigen in 2001, is a member of the B7 family [4] . It has been extensively studied in recent years as an important immune checkpoint molecule in the pathogenesis of tumors [5,6] . However, as an important immunomodulatory molecule, B7-H3 also plays a signi cant role in the pathogenesis of autoimmune diseases [7,8] . Some studies demonstrated B7-H3 as a positive costimulatory molecule and an IFN-γ-inducer in activated human T cells. Other experiments showed that B7-H3 transfectants could down-regulate T-cell proliferation and IFN-γproduction, suggesting that it also has inhibitory effect [9] . The cellular role of B7-H3 remains controversial until now [10] . Earlier studies, including our previous study, have shown that B7-H3 is associated with autoimmune diseases, such as RA, myasthenia gravis, ankylosing spondylitis, and systemic lupus erythematosus [11][12][13][14] . The expression of B7-H3 on peripheral blood monocytes of RA patients is closely related to clinical manifestations, although the mechanism of action of B7-H3 in RA remains unclear. In the present study, we analyzed (i) the expression of B7-H3 on macrophages in the joint cavities of RA patients and a collagen-induced arthritis (CIA) mouse model and (ii) the correlation of B7-H3 levels with the clinicohistological evaluation of the joint cavity. We studied the effects of reduced B7-H3 expression on the cellular in ammatory response and related signaling pathways through in vitro experiments, and we treated CIA mice with anti-B7-H3 in vivo to explore the possible role of this molecule in RA treatment.

Patients and controls
A total of 14 RA patients (4 males and 10 females) who underwent arthroplasty at the Department of Orthopedics, the First A liated Hospital of Suzhou University, Jiangsu Province, China, with a mean age of 54.7 ± 12.9 years and a disease course of 5-30 years were included in this study. Six patients (all males, mean age of 25.3 ± 4.1 years) with meniscus injury who underwent arthroscopic surgery in our hospital during the same period were selected as the control group. This study was approved by the Ethics Committee of Suzhou Health Vocational and Technology College. All patients signed written informed consent before participating.

CIA mouse model
The CIA mouse model was established as described in the study of Courtenay et al. [12] . Speci c pathogen-free DBA/1 male mice (7-8 weeks old, weighing 18-25 g) were purchased from the Shanghai SLAC Laboratory Animal Co., Ltd. (Shanghai, China). Mice were immunized (day 0) intradermally at the base of the tail with 100 μg/100 μl bovine collagen type II (Chondrex Inc., Redmond, WA) and mixed with 500 μg/100 μl of complete Freund's adjuvant (CFA) (Sigma-Aldrich).On day 21, after priming, the mice received an intradermal booster injection with the same mixture of CII with CFA. Mice were examined visually three times per week by the double-blind method, and the different appearance of peripheral joints was observed. Morbidity of the animals was observed and evaluated according to the following scoring system: score 0, normal; score 1, mild swelling and/or erythema; score 2, extensive swelling and/or erythema; and score 3, joint deformation and/or stiffness. Scores from all four paws were added to give the total for each animal.
2.3 Immunohistochemistry and immuno uorescence double staining of synovial tissue in RA patients and CIA mice The tissues obtained from replacement surgery of RA patients were xed in formalin and para nembedded. CIA mice were sacri ced a week after the onset of typical clinical symptoms including swollen or deformity ankles, the knee joints of the hind limbs were cut with bone scissors, and the surrounding tissues were removed. The joint samples were further xed in PBS containing 10% formalin, followed by decalci cation of the joints with 10% ethylenediaminetetraacetic acid (EDTA). The decalci ed joint samples were embedded in para n, sectioned, and stained with hematoxylin and eosin (HE) and immunohistochemistry. The pathological scores of the synovial joints, including the degree of cartilage destruction of the joint, pannus formation, and the in ltration of in ammatory cells (semi-quantitative scoring system, from score 0 to 3 for each item), were determined by a pathologist. The in ltration of in ammatory cells was scored with the following system: score 0, no in ammatory cells; score 1, sparsely scattered in ltration of in ammatory cells; score 2, relatively dense in ltration of in ammatory cells; and score 3, high degree of in ltration of in ammatory cells. Pannus formation was scored as follows: score 0, no formation; score 1, little (<50%) pannus formation; score 2, >50% of pannus formation; and score 3, very much pannus formation. The destruction of articular cartilage was scored as follows: score 0, no damage; score 1, small focal damage; score 2, 50% damage; score 3, 66.7% damage; and score 4, complete damage. The tissue sections were heated at high pressure for antigen retrieval, followed by the detection of the macrophage surface markers B7-H3 and CD68 by immunohistochemistry. Tissue sections were independently incubated with rabbit anti-B7-H3 (CD276, 1:100, Abcam) and anti-CD68 (clone KP1, Fuzhou Maixin Biology, Fujian Province, China) (80 μl each) at 37°C for 1 h, followed by washing in PBS three times (5 min each) and incubation with secondary antibodies (50 μl) at 37°C for 30 min. After three more washes in PBS, the tissue sections were stained with 3,3-diaminobenzidine (DAB) and subsequently hematoxylin, and the slides were mounted. The stained sections were observed under an Olympus optical microscope.
Fluorescence immunohistochemistry using tyramide signal ampli cation (TSA) technology was adopted in this study using a tyramide-coumarin conjugate kit (PerkinElmer, Waltham, MA), in accordance with the manufacturer's instructions. After adding the primary antibodies, the tyramide-coumarin working solution was added and samples were incubated for 15 min, followed by myosin immuno uorescence labeling using an indirect immuno uorescence method [5] . Polyclonal rabbit anti-human myosin IgG (1:20) was added and incubated at 37°C for 1 h, followed by three washes in PBS. The sections were then incubated with donkey anti-rabbit IgG-uorescein (FITC) (1:100) at room temperature in the dark for 1 h, followed by three washes in PBS (5 min each). Propidium iodide (PI) nuclear staining was performed by adding 1 μg/ml PI in PBS to the tissue sections and incubating at room temperature in the dark for 3 min.
After three washes in PBS at room temperature, the stained sections were observed under an AX-80 uorescence microscope (Olympus) with three channels: the FITC signal was observed in the green channel, the coumarin signal was observed in the blue channel, and the PI signal was observed in the red channel. Images were processed using Advance SPOT software.

Cell culture and viral transfection
THP-1(monocytes from human) and RAW264.7 cells (macrophage from mouse) were purchased from Sangon Biotech Co., Ltd. (Shanghai, China). The cells were cultured in RPMI 1640 medium containing 10% FBS and 100 U/ml penicillin-streptomycin at 37°C in a humidi ed incubator with 5% CO 2 . The viral LV3 vector used to construct a THP-1 cell line with stable low expression of B7-H3 contained shRNA against B7-H3 and was sequenced in our previous study [16] . The sequence complementary to the shRNA against human B7-H3 used in the construction of the THP-1 stable low expression cell line started at nucleotide 1,041 of the B7-H3 sequence (LV3-hB7-H3-1041). The sequence complementary to the shRNA against mouse B7-H3 used in the construction of the RAW264.7 stable low expression cell line started at nucleotide 294 of the B7-H3 sequence (LV3-mB7-H3-294). A universal interference sequence was used as the negative control (NC). The sequences and the viral vector were synthesized by Jikai Gene Chemical Technology Co., Ltd. (Shanghai, China) ( Table 1).
THP-1 and RAW264.7 cells in the exponential growth phase were seeded in 24-well plates at 5 × 10 4 cells/well and were incubated in an incubator containing 5% CO 2 at 37°C overnight. Culture medium was refreshed, 5 μl viral stock solution was added, which was successfully packaged in the 1.2.1, and 5 μg/ml nal concentration of polybrene infection/transfection reagent was added. After incubation 24h, the media was removed and replaced with 5 ml of fresh growth media containing 2 μg/ml puromycin to select for stable transfectants. Single clones were picked, and the B7-H3 expression was determined using RT-PCR and Western blot analysis.

Quantitative real-time PCR and western blot
Total RNA was puri ed from THP-1 or RAW264.7 cell line using Trizol (Takara Bio Inc., Shiga, Japan).). RNA was reverse-transcribed into cDNA with an Oligo(dT) primer using a PrimeScript First cDNA synthesis kit(Takara). qRT-PCR was performed on a 7500 Real Time PCR System (Applied Biosystems) in triplicate for each target transcript using SYBRGreen PCR Master Mix (Life Technologies). Relative mRNA expression was calculated with the comparative cycle threshold method. As a calibrator, the sample with the median cycle threshold value was used and set to 100%. The following primers for human B7-H3 were used: forward primer 5′-CTTGTTCGATGTTCACAGCG-3′ and reverse primer 5′-GCCGTAGAGCTGTCTTGGATC-3′. Mouse B7-H3 primer was: forward primer 5′-TCAGTCACCATCACAGGGCA -3′ and reverse primer 5′-GAGGGTTTCAGAGGCCGTAG -3′ Cell lysates (40 mg) were subjected to SDS-PAGE, then transferred electrophorectically onto polyvinyldi uoride (PVDF) membranes (Millipore, Bedford,MA). Blots were blocked with 4% BSA for 1 h at room temperature,then probed with primary antibodies against human B7-H3(AF1027,R&D Minneapolis,MN,USA) and Mouse B7-H3(ab134161, Abcam,cambridge,MA,USA) for another hour at room temperature. After three washes, blots were subsequently incubated with the horseradish peroxidaseconjugated secondary antibodies for 1 h at room temperature and visualized by enhanced chemiluminescence using ImageQuant™ LAS 4000 (GE Healthcare, Pewaukee, WI).

Establishment of an in vitro in ammation model
THP-1 cells in the exponential growth phase were seeded in six-well plates at a concentration of 1 × 10 6 cells/ml. After 24 h, PMA (at a concentration of 100ng/ml) was added to the cells. After 48 h of stimulation, the supernatants were collected and stored in a refrigerator until further analysis. The cell pellets were lysed, followed by protein extraction, and the extracted protein samples were stored at −80°C until further analysis. The RAW264.7 cells were also seeded in six-well plates at a concentration of 1 × 10 6 cells/ml. After 24 h, LPS (at a concentration of 100ng/ml) was added to stimulate the cells for 48 h, and the supernatants and the cell pellets were collected. Human-TNF-a-Forward:5'-CTTCTCGAACCCCGAGTGAC-3' Human-TNF-a-Reverse:5'-ATGAGGTACAGGCCCTCTGA-3' The serum levels of the cytokines interleukin (IL)-6, TNF-α, and IL-1 were quanti ed using a LegendPlex TM Multiplex cytokine bead-based assay (BioLegend, San Diego, CA) following the manufacturer's instructions. For each reaction, 25 µl of serum was diluted with 125 µl deionized water. All samples were run on an Accuri C6 ow cytometer (BD Biosciences, San Diego, CA) and acquired data were analyzed using LegendPlex v7.0 software.

NF-κB signaling
The NF-κB Pathway Sampler Kit (catalog number: 9936, Cell Signal Technology, Danvers, MA) was used for the detection of NF-κB pathway-related proteins. Cultured cells were harvested and proteins were extracted. The levels of NF-κB p65, p-NF-κB p65, IκBα, and p-IκBα were measured by western blot.

Treatment of CIA mice with anti-B7-H3
CIA mice began to develop disease 25 days after the booster immunization. Twelve randomly selected mice were injected with monoclonal anti-B7-H3 (catalog number: 135604, Biolend) via the tail vein. Each control mouse was injected with 100 µg rat IgG2a, k isotype (catalog number: 400533, Biolend) once every other day for a total of 9 days. before the beginning of the treatment and at the end of the treatment, 0.5 ml blood sample was collected using a submandibular vein sampling needle from the submandibular vein plexus. The serum was separated and the expression levels of TNF-α, IL-6, and IL-1β were measured by ow cytometry. In addition, the joints of animals were evaluated and scored before and after treatment. The mice were sacri ced to dissect the joints for xation and para n embedding. Tissue sectioning and staining were performed as described above, and the histological evaluation of mice was performed by a pathologist.

Statistical analysis
All calculations were performed with SPSS 17.0 (IBM SPSS Inc., Chicago, IL). Data are expressed as mean ± SD. Statistical analysis was performed with the Student t test and ANOVA. Correlations were evaluated by the Pearson correlation test. The r 2 value was calculated to evaluate linkage disequilibrium. P <0.05 was considered a statistically signi cant difference.

Page 8/23
3.1 B7-H3 was highly expressed on macrophages in synovial tissue of RA patients and CIA mice Pathological changes of synovial joint tissues were detected by HE staining. The joint tissue slices of the control group showed a clear joint cavity. The synovial cells were neatly arranged, and the nonin ammatory cells in ltrated in the synovial tissue. The surface of the articular cartilage was smooth. In the experimental groups, the joint tissues showed hyperplasia of synovial lining layers, both in RA patients and CIA mice, showing more than 10 synovial lining layers. The chondrocytes were arranged disorderly, and the joint space disappeared or even merged. In addition, the tissues showed invasion of immune cells, including neutrophils, T cells, and macrophages. Tissues of some patients and mice had cartilage and synovial tissues at the joint tissues ( Fig. 1A and B).
We also analyzed the expression of B7-H3 in the joint tissues of 16 RA patients and 69 CIA mice. Immunohistochemical analysis revealed that B7-H3 was expressed in synovial broblasts, macrophages, and other cells in the synovial tissues of RA patients and CIA mice. To further investigate whether B7-H3 was expressed in macrophages, multiple immuno uorescence analyses showed that B7-H3 and CD68 were coexpressed in the macrophages of CIA mouse joint tissues (Fig. 1C). In addition, we labeled macrophages in the tissues with anti-CD68 to analyze the in ltration of macrophages. Immunohistochemical analysis showed that the number of in ltrating macrophages in CIA mice was signi cantly higher than in control mice ( Fig. 2A). The number of in ltrating macrophages was closely related to the joint score of the mice as evaluated by the level of synovial TNF-α (Fig. 2B) and histological analysis, including the degree of cartilage destruction and the level of angiogenesis ( Fig. 2C and D). Immunohistochemical analysis showed that B7-H3 expression levels on macrophages were closely related to the joint score (r 2 = 0.4131, P < 0.0001, Fig. 2B) and to the histopathological evaluations, including cartilage destruction and pannus formation (P < 0.001).
The semi-quantitative immunohistochemical analysis of TNF-α expression in the joint cavity showed TNF-α was diffusely distributed in the cytoplasm of interstitial cells of joint tissues (Fig. 4A and B). TNF-α levels in the peripheral serum were measured by ELISA. Correlation analysis showed that higher numbers of B7-H3-positive macrophages in the joint cavity tissues were associated with stronger expression of TNF-α in joint tissues (P < 0.0005; Fig. 4C) and with higher levels of TNF-α in peripheral blood (P = 0.0091; Fig. 4D).

Stable knockdown of B7-H3 by shRNA in THP-1 and RAW264.7 macrophage cell lines dampens proin ammatory cytokine production upon stimulation
To establish the in vitro in ammation cell models, 100 ng/ml PMA and 100 ng/mL LPS were used to activate human THP-1 and mouse RAW264.7 macrophage cell lines [17,18], respectively, for 48 h. The culture medium was collected and the levels of TNF-α, IL-6, and IL-1 were measured using a multicytokine detection system.
To test the role of B7-H3 in in ammatory responses of human and mouse macrophages, we knocked down B7-H3 expression in THP-1 and RAW264.7 macrophages and measured their pro-in ammatory cytokine production upon stimulation. The LV3-hB7-H3-1041 viral vector prepared in our previous study was used in THP-1 cells, and the LV3-mB7-H3-294 viral vector, designed to target the sequence following nucleotide 294 in mouse B7-H3, was used in RAW264.7 cells. Real-time PCR showed than when THP-1 cells were transfected with LV3-hB7-H3-1041, the B7-H3 mRNA levels were reduced by 67.2% compared with controls (Fig. 5A). When RAW264.7 cells were transfected with LV3-mB7-H3-294, B7-H3 mRNA levels were reduced by 72.6% compared with controls. Western blot analysis con rmed that the B7-H3 protein levels were decreased in both cell lines (Fig. 5B-D).
After B7-H3 knockdown, the levels of TNF-α and IL-6 were signi cantly lower relative to the control group ( Fig. 6A and C, P < 0.01), while no signi cant changes in IL-1 levels were observed (not shown). In addition, cells were harvested and the TNF-α and IL-6 mRNA levels were measured by PCR. Both were signi cantly downregulated ( Fig. 6B and D).

B7-H3 regulated the in ammatory response in macrophages through the NF-κB pathway
Because NF-κB signaling regulates the transcription and expression of various in ammatory responserelated genes, it is closely related to the occurrence and maintenance of in ammation. To further verify whether NF-κB is involved in the B7-H3-mediated regulation of in ammation in macrophages, we analyzed the levels of the NF-κB signaling-related proteins NF-κB p65, p-NF-κB p65, IκBα, and p-IκBα, in THP-1 cells after 48 h PMA stimulation and RAW264.7 cells after 48 h LPS stimulation by western blot analysis. Upon B7-H3 downregulation, the protein levels of NF-κB p65, p-NF-κB p65, IκBα, and p-IκBα were signi cantly decreased compared with the control group (LV-NC) (Fig. 7).

Anti-B7-H3 alleviated the clinical manifestations and in ammation of arthritis in CIA mice
On day 7 after symptom onset in CIA mice, monoclonal anti-B7-H3 was injected into the tail vein once every other day for 9 days. The joint scores were taken and blood was drawn from the submandibular vein before and after Ab therapy. In CIA mice receiving antibody treatment, the joint swelling was signi cantly reduced (Fig. 8A); the second and third toes of the mice showed the most obvious reductions (Fig. 8A). Nine CIA mice were treated with anti-B7-H3, and two of them died during treatment. Six of the remaining seven mice had a decreased joint score (Fig. 8B). After anti-B7-H3 treatment, the expression of the pro-in ammatory cytokines TNF-α, IL-1β, and IL-6 in ve CIA mice was signi cantly reduced compared with before treatment (Fig. 8C). In addition, histopathological evaluation of the articular cartilage destruction, in ammatory cell in ltration, and pannus formation of the antibody-treated group showed these were signi cantly alleviated compared with the control group.

Discussion
In ammation-related cells and mediators play important roles in RA. One of the most important features of RA is the sharp increase in in ammation-related cells, such as macrophages, T cells, B cells, dendritic cells, and neutrophils, at the synovium of joints [19]. These in ammatory cells secrete large amounts of pro-in ammatory mediators and degradative enzymes, such as matrix metalloproteinases. These molecules mediate interactions between cells and promote the in ammatory response, which eventually leads to the destruction of the synovium, cartilage, and bone tissue [20]. Macrophages are an important class of in ammatory cells at the synovium of joints. Inactive peripheral blood monocytes mature and in ltrate the in amed tissues, where they differentiate into macrophages upon stimulation and regulation by pro-in ammatory cytokines [21,22]. In turn, they interact with activated T cells, broblasts, epithelial cells, and NK cells in the joints by releasing pro-in ammatory cytokines, such as TNF-α and IL-1, chemokines, and macrophage migration inhibitory factors [23]. In the present study, we observed signi cantly more in ltrated macrophages in the joint tissues of RA patients and CIA mice than in healthy individuals and control mice. In addition, the secretion of in ammatory cytokines in the tissues was high. The degree of macrophage in ltration was positively correlated with the secretion of in ammatory cytokines, such as TNF-α, and higher degrees of cartilage destruction and pannus formation were associated with higher degrees of macrophage in ltration.
In general, the protein levels of B7-H3 are relatively limited compared with the mRNA levels [4] . Except for malignant tumor tissues, different degrees of B7-H3 expression have only been detected on dendritic cells and macrophages in tissues and activated peripheral blood monocytes in vitro [24,25] . In a previous study we showed that the expression of B7-H3 on the surface of CD14 + monocytes in peripheral blood of RA patients is signi cantly higher than in healthy individuals, and is positively correlated with the clinical pathology and parameters, including the number of swollen joints, the DSA28 score index, the duration of disease, the erythrocyte sedimentation rate, and the levels of rheumatoid factors [14]. Another study has shown that B7-H3 is expressed in the synovial tissues of RA patients, CD68 + macrophages, synovial broblasts, CD31 + endothelial cells, and CK18 + epithelial cells [26]. In the present study, we analyzed B7-H3 expression on macrophages of synovial tissues from joints of RA patients and CIA mice by immunohistochemical analysis and immuno uorescence costaining, and observed that B7-H3 expression in these cells was signi cantly higher than on the macrophages of normal synovial tissues. Furthermore, we demonstrated that B7-H3 expression levels were positively correlated with the clinical and histological evaluations, including joint score, cartilage destruction, and pannus formation, and were weakly correlated with the secretion of in ammatory cytokines. Most studies have shown that B7-H3 has a co-inhibitory role on immune response, contributing to tumor cell immune evasion Targeting B7-H3 in cancer treatment can reduce cell proliferation, progression, and metastasis, which may ultimately lead to improved therapeutic options and better clinical outcomes. Our results novelty indicated B7-H3 might act a co-stimulate role in immune response and related with in ammation-related autoimmune disorders.
What role does high B7-H3 expression levels on macrophages in joint tissue play in RA? B7-H3 may be related to the in ammatory regulation of macrophages. A study has shown that B7-H3 signi cantly enhances the toll-like receptor (TLR)-4 agonist LPS and the TLR-2 agonist bacterial lipoprotein to stimulate the production of pro-in ammatory cytokines in monocytes and macrophages, regulating the in ammatory response; MIH35 (an anti-B7-H3 antibody) signi cantly reduces the release of proin ammatory cytokines [27]. Chen et al. have shown that B7-H3 activates TLR-2 downstream signaling pathways, NF-κB p65, and MAPK p38 in mice infected with Streptococcus pneumoniae. Inhibitors of the above pathways weaken the pro-in ammatory responses and reduce the production of pro-in ammatory cytokines and chemokines [28]. To further explore the effects of high B7-H3 expression on macrophages, we constructed THP-1 cells (human mononuclear phagocytes) and RAW264.7 cells (mouse macrophages) with stable low B7-H3 expression, and showed that these stable B7-H3 knockdown cell lines had reduced secretion of pro-in ammatory cytokines, including IL-6 and TNF-α, under an in ammatory environment. In addition, signaling pathway analysis showed that NF-κB signaling was inhibited. We con rmed that B7-H3 may enhance the ability of macrophages to regulate in ammation through the NF-κB signaling pathway and participate in RA progression.
To verify our hypothesis, we treated CIA mice with anti-B7-H3. Moreover, the levels of the proin ammatory cytokines IL-6 and TNF-α in the peripheral blood were signi cantly lower, and the expression of in ammatory cytokines, the degree of in ammatory cell in ltration, and the degree of bone and joint destruction were signi cantly reduced. Based on our ndings, anti-B7-H3 treatment may reduce in ammation and mitigate RA symptoms and damage.

Conclusions
In this study, we have shown that high expression levels of B7-H3 in macrophages in RA tissues enhance the ability of macrophages to promote the in ammatory response. Blockade with anti-B7-H3 reduced the in ammatory response, suggesting that B7-H3 may be used as a drug target for the treatment of RA.

Consent for publication
Not applicable

Availability of data and materials
The data that support the ndings of this study are available on reasonable request from the corresponding author.

Competing interests
The authors declare that they have no competing interests.    blot analysis con rmed low B7-H3 protein expression in RAW264.7 cells transfected with LV3-mB7-H3-294. Data are presented as means ± SD from three independent experiments.   (B) Anti-B7-H3 signi cantly reduced the TNF-α level in peripheral blood after anti-B7-H3 treatment for 9 days. (C) Anti-B7-H3-treated DBA/1 mice induced with CIA for 9 days show a reduction in clinical arthritic score compared to isotype IgG control. (D) TNF-α, IL-6, and IL-1 levels in seven CIA mice before and after anti-B7-H3 treatment. n = 9/normal group, n =7-9 /isotype IgG control and B7-H3 Ab groups.