The Role of CCL3 in the Pathogenesis of Rheumatoid Arthritis

Chemokines, also known as chemotactic hormones or chemotactic agents, are a small molecule type of cytokine that induces chemotaxis in cells [12]. The structure and function of chemokines have been further clarified. Chemokines have a highly conserved tertiary structure defined by an irregular “N-loop”, a C-terminal helix, and a three-stranded β-sheet connected by loops. So far, more than 40 human chemokines have been found, belonging to the largest family of cytokines. To facilitate communication, since the late 1990s the chemokine international naming organization has mainly been divided into four chemokine subfamilies according to the relative differences of cysteine positions in the structure [13]. There are three other arbitrary amino acids between two cysteines at the N-terminal of the CX3C chemokine subgroup, one other arbitrary amino acid between two cysteines at the N-terminal of the CXC chemokine subgroup, and two cysteines at the N-terminal of CC chemokine subgroup are adjacent to each other. Most chemokines have four conserved cysteine residues. However, the XC chemokine subfamily has only two cysteine residues [14] (Fig. 1). The XC chemokine subfamily contains only XCL1 and XCL2 chemokines. The main function of chemokines is chemotactic immune cells involved in the body's defense and exudation of inflammatory process [15].

There are currently more than 20 chemokine receptors. According to the difference of binding ligands, they divide into the following four types: CXCR, CCR, CX3CR, and XCR. Chemokine receptors belong to the G-protein-coupled receptor superfamily of seven transmembrane receptors. They share many structural features: similar in size with approximately 350 amino acids. There is homology between different chemokine receptors [16]. Studies have shown that chemokine receptor N-terminal amino acid residues deletion cannot bind ligands. This confirms that the extracellular N-terminal sequence is the recognition site for ligands. The chemokine receptor C-terminal contains Ser/Thr site, which is mainly involved in intracellular signal transduction, and C-terminus phosphorylation may be involved in signal transduction after receptor activation. Ligand-receptor binding of chemokines is a classical two-site model [17]. In the first step, the chemokine core domain binds to the N-terminal recognition site of the chemokine recognition site 1 (CRS1). The core region of chemokines refers to structures other than N-terminal sequences. Then, the N-terminal of chemokine binds to the transmembrane domain of chemokine recognition site 2 (CRS2) (Fig. 2).

In the process of chemokine ligand–receptor binding, we can see the redundant characteristics of the binding. It determines the specificity of chemokine action with its receptor. Therefore, chemokines have different expressions in synovial fluid of patients with arthritis [18]. In the family of chemokines, several chemokines are involved in the pathogenesis of RA (Table1). Macrophages secrete CXCL1, CXCL5, and CXCL8, which trigger synovial inflammation and fibrosis [19‐21]. A new study indicates that that CXCL7 levels were increased in RA patient platelet [22]. Neutrophils produce CXCL6 and trigger the production of CXCL4, CXCL9, CXCL10, and CXCL16 from fibroblast-like synovial cells and macrophages [23‐25]. Furthermore, CXCL12 stimulated osteoclast migration and bone resorption of osteoclasts in RA animal models [26]. Elevated levels of CCL2, CCL3, and CCL5 in RA were closely associated with increased T cell infiltration in joints [27]. Studies have shown that CCL13 is correlated with RA, and CCL13 is highly expressed in chondrocytes [28]. Besides, CCL18 promotes the attraction of T cells to joint inflammation through antigen-presenting cells [29, 30]. XCL1 and XCL2 synergistically up-regulated CD4⁺/CD28⁻ T cells, thereby inducing osteoclast proliferation [31, 32]. CX3CR1 expressed by monocytes interacts with CX3CL1 produced by macrophages, and this interaction promotes their migration and recruitment into RA synovium [33].

The synovial tissue and synovial fluid of RA contain plentiful chemokines, and their receptors have higher expression in T cells, monocytes, neutrophils, and other immune cells. In general, a variety of chemokines derived from immune cells can cause the immune activation of RA, and it is of great significance to clarify their mechanism of action for the prevention and treatment of RA, especially representative inflammatory chemokines such as CCL3 and CCL2.

Inflammatory cytokine infiltration in joint synovium is one of the main features of RA disease [37]. Zhang et al. demonstrated inflammatory cell status in synovial tissues of RA by integrating single-cell transcriptome and mass cytography. For example, in leukocyte-rich RA, CCL3 is mainly upregulated in monocytes. However, in RA and osteoarthritis (OA) synovium without leukocytes, the CCL3 expression level was comparable in B and T cells [38]. Using immunofluorescence technology to further analyze the localization of CCL3 in synovial tissues, the results show that CCL3 was highly expressed in the cytoplasm of CD4⁺ T cells. When synovial macrophages and fibroblasts are stimulated by pro-inflammatory factors, synovial cells are activated to produce CCL3. More importantly, CCL3 recruits more inflammatory cells to migrate and accumulate in the joint, then inflammatory cells generate more inflammatory factors in the synovial microenvironment, further aggravating the local inflammatory response in RA [39]. In rheumatoid arthritis fibroblast-like synovial cells (RA-FLS), Zhang et al. observed that CCL3 causes the onset of RA through PI3K/AKT signaling pathway. Besides, CCL3 could active CD4⁺ T cells to mediate synovitis [40]. CCL3/CCR5 may overexpress inflammatory mediators such as cyclooxygenase-2 (COX-2), matrix metalloproteinases-3 (MMP-3), and interleukin-6 (IL-6) in synovial cells through mitogen-activated protein kinase pathway, thus contributing to RA synovitis [41]. In RASF, mRNA sequencing data showed that Mir-147b-3p mimics and ZNF148 siRNA usually upregulate CCL3 and other inflammatory pathways mediate synovitis [42]. All of the above studies indicate that CCL3 is involved in the pathological process of synovitis.

Cartilage destruction involves an interaction between osteoclast (OC) generation and osteoblast (OB) inhibition. Joint bone destruction is an irreversible process and is the main cause of disability in patients. Studies have shown that many cytokines and immune cells contribute to osteoclast formation [43, 44]. Immune cells in the pathogenesis of RA directly up-regulate the expression of receptor activator of nuclear factor-kB Ligand (rankl), but indirectly increase the expression of rankl through secretion of pro-inflammatory cytokines, and promote the formation of OCs, ultimately leading to cartilage fragmentation [45]. CCL3 is a chemokine of macrophages, which can induce osteoclast aggregation and activity [46]. Various rodent studies in vitro have demonstrated a link between CCL3 production and bone degradation [47, 48]. Studies have shown that CCL3 can promote the differentiation of OC precursors in peripheral blood. It is worth noting that curcumin further inhibits the differentiation of osteoclast precursors and the formation of osteoclasts by inhibiting the production of CCL3. The addition of CCL3 could significantly reverse the OC formation decreased by curcumin [49]. Zhang et al. demonstrated that C/EBPβ and NF-κB are involved in the expression of CCL3 in IL-1β-induced human chondrocytes [50]. In TNF-Tg and CIA mouse models and in human RA samples, Wen Sun et al. showed that B cells in synovial tissue significantly reduced osteoblast differentiation through the high expression of CCL3 [10]. In addition, most studies have confirmed that CCL3 is an inhibitor of OBs [51]. These findings suggest that CCL3 is a major contributor to cartilage matrix degradation in RA. In the CIA model, CCL3^−/− mice were protected from bone injury, further demonstrating the important role of CCL3 in osteoclasts [34]. However, CCL3 is indirectly involved in the pathological process of bone resorption in osteoclast. MMPs lead to bone degradation by dissolving the lacunar organic matrix. Studies have shown that the increased expression of CCL3 and enhanced activation of MMPs in human CD14 OC promote the differentiation and reabsorption of OC [52]. Most importantly, other studies have shown that CCL3 increases MMP3 activity in FLS cells in a culture medium in a dose-dependent or time-dependent manner. MMPs can promote the migration of FLS cells to adjacent cartilage and induce cartilage degradation [53]. Therefore, CCL3 may play an important role in RA cartilage degradation through autocrine or paracrine action.

Pannus is composed of newly neovascularization, proliferating synovial cells, inflammatory cells, and cellulose, with tumor-like features [54]. The proliferation of synovial vessels may be one of the reasons why RA is difficult to cure, a feature similar to vascular proliferation in malignant tumors. Angiogenesis is a vital factor in the formation and maintenance of RA pannus. On the one hand, the formation of neovascularization will increase the surface area of vascular endothelium, which makes it easier for inflammatory cells to seep into the synovium, thus aggravating inflammation. On the other hand, some mediators that regulate angiogenesis are also involved in the pathogenic mechanism of RA [55]. Arthritis often causes local hypoxia due to active metabolism in the joints. In the local hypoxic microenvironment of the joint during RA, synovial fibroblasts secrete CXCL12. It combines with its receptor CXCR4 to participate in vascular remodeling and promote the migration of various endothelial cells, thereby promoting capillary formation. The new vessels recruit more immune inflammatory cells into synovial tissue with the help of inflammatory chemokines such as CCL3 [56]. In addition, Liao et al.’s studies have shown that CCL3 inhibits mir-374b expression by activating the JNK/ERK/P38 signaling pathway, thereby promoting VEGF-A expression and angiogenesis in human osteosarcoma cells [57]. After being stimulated by some proinflammatory cytokines, synovial fibroblasts in RA joint tissue secrete CCL3 and CCL28, which are also involved in vascular proliferation in RA. Their promotion of vascular proliferation is mainly indirectly by macrophages and fibroblasts in synovial tissue to secrete pro-angiogenic factors [58‐60]. In a mouse CIA model, Krausz et al. found that the synergistic effect of angiopoietin-2 (Ang-2) and TNF-α on Tie2 signaling in macrophages enhances the expression of CCL3 and IL-6 while inhibiting the production of thrombospondin-2 (TSP-2), widely coordinating angiogenesis in RA [61]. Therefore, blocking angiogenesis will become a new treatment for RA.

RA is an immunological, systemic disease with clinical manifestations involving joints and other systemic systems. For example, interstitial lung diseases are the most common, which can eventually progress to pulmonary fibrosis [62]. Through radiation lung injury irradiation on a CCL3^−/− mouse model, CCR1 ^−/− mouse model, and wild mice, it was found that the degree of pulmonary fibrosis in normal mice was significantly higher than that in CCL3^−/− mice and CCR1^−/− mice. CCL3 can initiate NF-kB signaling to participate in the pathogenesis of pulmonary interstitial lesions. Therefore, selective blocking of the CCL3/CCR1 signaling axis can effectively relieve pulmonary fibrosis secondary to RA [63]. TGF-β1 is a typical cytokine-stimulating epithelial–mesenchymal transformation. It may stimulate CCL3 expression in lung NK cells and is associated with increased TGF-β1 in RA patients. Inhibition of TGF-β signaling can down-regulate pulmonary CCL3 to regulate interstitial pneumonia [64]. These studies show that CCL3 may be involved in the progression of lung diseases and is most closely related to the formation of pulmonary fibrosis.