In the early stages of infection, host cells secrete large amounts of chemokines that recruit neutrophils to the infectious site and exert their bactericidal effects. CHIPs and FPRL1 inhibitory protein (FLIPr) are two virulence factors secreted by S. aureus which affect neutrophils’ chemotaxis function in osteomyelitis. Chemotaxis inhibitory proteins (CHIPs) reduce chemokine production and reduce neutrophil chemotaxis toward the infectious site (12). FLIPr inhibits formyl peptide receptor-like-1 (FPRL1) by decreasing intracellular calcium mobilization and facilitates neutrophil chemotaxis to the infectious site. In this condition, neutrophils are unable to arrive at the infectious site and thus cannot kill pathogens through reactive oxygen species and NETs (13). Thus, the reduced chemotactic capacity of neutrophils leads to a much reduced bactericidal effect of innate immunity. Therefore, targeting CHIPs and FLIPr secreted by S. aureus would be a new strategy to enhance neutrophil chemotaxis.

It has been well known that the life span of neutrophils is prolonged in patients with sepsis or severe bacterial infections (14), and Wagner et al. demonstrated that these prolonged neutrophils still produce superoxide and are equipped phagocytic activity, but their chemotaxis is significantly reduced (15). The team later showed that the neutrophil expression of the activation-related surface receptors including CD14 and CD64 remained upregulated during persistent infection, while the chemoattraction-related adhesion molecule CD62L was reduced, which may explain the lack of neutrophil chemotaxis in persistent infection (16).

In general, neutrophil activation depends on the recognition of pathogen surface receptors. Staphylococcal superantigen-like 3 (Ssl3) belongs to the staphylococcal-like protein (Ssls) family and is also a virulence factor secreted by S. aureus, which blocks pathogen recognition by neutrophils and inhibits their activation by competitively binding to Toll-like receptor 2 (TLR2) (17, 18); the virulence factor CHIPs bind competitively to complement C5a, preventing C5a from activating neutrophils after binding to the complement receptor (3). As a response, neutrophils secret inhibitory neutrophil elastase proteinase-3 and cathepsin G to digest CHIPs. However, S. aureus is able to produce extracellular adherence protein (Eap) and Eap homologs (EapH1 and EapH2), which, in turn, inhibit these neutrophil elastases (19). Moreover, during persistent infection, some neutrophils express MHC class II, which is able to present antigen to T cells. It has been demonstrated that such neutrophils in chronic infection have the capacity to present antigen and activate T cells (16).

A distinctive feature of S. aureus in osteomyelitis is the ability to form biofilms on the surface of the implant. Generally, without the shelter of biofilm, planktonic bacteria are quickly recognized and captured by the immune cells. Biofilms are formed when planktonic bacteria adhere to the surface of implants (e.g., internal fixations: plates, screws, etc.) (5). Immature biofilms cannot serve as a perfect barrier and can be penetrated by neutrophils, which are fully activated to perform their immune function. However, once the biofilm is mature, its function as a physical barrier to immune cells is greatly enhanced. Extracellular polymers (EPS) are a major component of biofilms and are the main immunosuppressive effector. EPS impede the recognition of bacteria by shifting the recognition target of neutrophils from the pathogen to itself, thus helping the pathogen to evade the host’s immune response and persist at the infectious site (20, 21). However, extracellular DNA (eDNA), one of the components of EPS, has the ability to activate neutrophils. It is most likely that the activation of neutrophils by eDNA is not dependent on TLR; surface molecules of neutrophils are possibly able to directly sense eDNA in the bacterial microenvironment and are then activated (22). It is clear that this important component of EPS has a completely different effect on immune cells than EPS. However, whether there are other components of EPS that affect immune cell function remains to be investigated in the future.

Neutrophils are the first line of defense for innate immunity and are usually involved in the immune response through phagocytosis and internalization, activation of degradative enzymes and cationic peptides in granules, extracellular trap formation, and antigen presentation (16, 23).

Staphylococcal complement inhibitor (SCIN) secreted by S. aureus suppresses neutrophil activation by inhibiting C3 convertase activity, which results in reduced complement C3 production as well as suppressed opsonization and phagocytosis in neutrophils (24, 25). As another virulence factor, Ssls7 inhibits the function of chemotaxis and complements production, pathogen phagocytosis and reactive oxygen species production (26, 27). Besides that, many other virulence factors are also involved in the inhibition of phagocytosis and killing of neutrophils. S. aureus protein A (SpA) inhibits antigen phagocytosis; alpha-type phenol-soluble modulin (PSMα), pantone viren albumin (PVL), albumin GH (LukGH), and gamma hemolysin (Hlg) induce neutrophil lysis and apoptosis and reduce the number of neutrophils (3).

The persistent intracellular survival of S. aureus is a crucial reason why neutrophils cannot play a full role in phagocytosis and killing. S. aureus interferes with the function of the neutrophil phagosome and thus persists in immune cells. The virulence factors superoxide dismutase (SodA and SodM), catalase (KatA), alkyl peroxide reductase (AhpCF), and streptomycin (encoded by crtOPQMN) modulate the cytotoxic effects of ROS, allowing pathogens to persist in the phagosome and preventing their complete elimination (28–30); the virulence factor Hla attenuates ROS-mediated killing by blocking the recruitment of phagosome toward the mitochondria. In recent years, Neumann et al. have found that S. aureus was usually labeled and killed in LC3-positive phagosomes, whereas S. aureus cannot be a target in phagosomes without LC3 modification. Therefore, in osteomyelitis, it is likely that S. aureus survives in the phagosome by interfering with the expression of LC3-modified protein production, and it is necessary to investigate the mechanisms by which S. aureus affects LC3 expression in osteomyelitis; it may be a new strategy to effectively block these pathways and restore the function of the phagosome to kill pathogens (31).

The formation of extracellular traps is crucial for neutrophils to clear pathogens. NETs are released by neutrophils and are covered with histones, elastase, myeloperoxidase, and cathepsin G (32). The function of NETs is to capture and clear S. aureus at the infectious site at the expense of inducing damage to endothelial cells and other organs (33). Fabrizio Semeraro et al. demonstrated that histone proteins mediated platelet activation through TLR4 and TLR2 to promote blood coagulation and thrombin formation. The researchers proposed that this may be related to platelet-rich microthrombosis in sepsis models. We speculated that it was possible that the histones in NETs also play a key role in the formation of microthrombosis in chronic osteomyelitis, which hinders tissue repair and provides conditions for the growth of S. aureus. Therefore, NETs not only trap pathogens but also play an important role in tissue destruction in chronic osteomyelitis (34).

In summary, S. aureus exerts a series of negative effects on neutrophils to escape immune reactions. Inflammatory factors like IFN-γ, EPS, and NETs are involved in such process, which made all of them become possible therapeutic targets. It is worth noting that virulence factors almost affect the overall process, especially in phagocytosis and killing. Therefore, how to clear virulence factors secreted by S. aureus remains to be investigated.

In the early stages of infection, circulating monocytes are differentiated into macrophages by inflammatory cytokines. M1 macrophage is able to kill pathogens by releasing lysosomal enzymes and promoting inflammation by secreting pro-inflammatory factors. M2 macrophage secretes anti-inflammatory factors to inhibit excessive inflammation and produces platelet-derived growth factor, fibroblast growth factor, etc., to promote tissue repair (35). In chronic osteomyelitis, the interconversion and percentage between the two subtypes of macrophages are abnormal and imbalanced. The persistent presence of M1 macrophages induces an excessive inflammatory response, which may lead to severe tissue destruction. Likewise, the excessive production of M2 macrophages leads to insufficient phagocytosis and cytotoxic effects, biofilm formation, and bacterial resistance (36). It indicates that the mechanisms of macrophage action in chronic osteomyelitis are complex.

The STAT3/STAT6 pathway promotes macrophage polarization toward the M2 phenotype, whereas the STAT1 pathway promotes macrophage differentiation to the M1 phenotype. This suggests that the STAT3/STAT6 pathway negatively mediates the pathogen clearance ability in macrophages, which induces the persistence of osteomyelitis. Furthermore, it has been shown that IL-10 can trigger the STAT3 pathway to suppress the immune response of macrophages. Therefore, targeting this signaling pathway in macrophage may be a new immunoregulatory therapy for treating chronic osteomyelitis (37–39).

The biofilm formed by S. aureus is also able to influence macrophage polarization. When biofilms are not yet formed, the metabolic pattern of host immune cells is mainly based on glycolysis, which is able to participate in macrophage-mediated pro-inflammatory responses and contribute to the conversion of macrophages to the M1 phenotype. In contrast, when biofilms are formed, bacterial populations are significantly higher; thus, glucose and nutrients in the microenvironment are barren. Inefficient energy-producing pathways force host immune cells to choose oxidative phosphorylation to support macrophage conversion to M2 phenotype to repair damaged tissues (40, 41). Therefore, targeting mature biofilms formed by S. aureus may be a new strategy to modulate the excessive inflammatory response and tissue destruction caused by an imbalance in the ratio of M1 and M2 phenotypes of macrophages in osteomyelitis.

Like neutrophils, macrophages are recruited by chemokines and leave the circulation to the infectious sites. In chronic osteomyelitis, macrophage migration is regulated by the NF-κB/TWIST1 signaling pathway. NF-κB induces the upregulation of TWIST1 expression to enhance macrophages polarized to M1 phenotype and migrate to the infectious site (36). In addition, TWIST1 induced the expression of MMP9 and MMP3, which can also promote macrophage migration (42).

Macrophages exert their immune effects mainly through pathogen phagocytosis. Macrophages are regulated by the PI3K/Akt-Beclin signaling pathway when they exert phagocytosis and killing under osteomyelitis conditions.

The PI3K/Akt-Beclin signaling pathway regulates macrophage autophagy, pathogen phagocytosis, and NF-kB-mediated inflammatory responses in S. aureus osteomyelitis. Recently, studies have shown that the PI3K inhibitors inhibited macrophage autophagy and impaired the phagocytosis of S. aureus. Meanwhile, when Beclin1 was knocked out and autophagy was inhibited, the NF-κB signaling pathway was activated, and a significant increase in inflammatory factors, such as TNF-α and IL-1β, can be detected in macrophages (43). It is clear that PI3K/Akt-Beclin positively regulates macrophage autophagy and phagocytosis and negatively regulates NF-κB-mediated TNF-α and IL-1β production in macrophages.

Obviously, some signaling pathways and relative transcription factors play key roles in macrophage differentiation, chemotaxis, and phagocytosis. It is important for us to explain the relationship among these signaling pathways, S. aureus, and macrophages. Mediating these pathways may become a new therapy in osteomyelitis treatment.

Bone formation, a dynamic process, is mediated by osteoblast-mediated bone formation and osteoclast-mediated bone resorption (20). Many studies have shown that macrophages were involved in bone formation and bone resorption in inflammation by regulating the functions of osteoblasts and osteoclasts. In chronic osteomyelitis, the balance of bone remodeling is disrupted, osteoblasts differentiation is inhibited, and more osteoclast precursor cells differentiate into osteoclasts, of which macrophages are significantly relevant (10, 44, 45).

Most dendritic cells are able to activate T cells by presenting antigens. Leukocidin AB (LukAB), a virulence factor secreted by S. aureus, prevents T cells from being activated by killing dendritic cells, which play a key role in antigen presentation (65).

Cytotoxic T lymphocyte A antigen 4 (CTLA-4) is secreted primarily by T cells and is an inhibitory molecule that binds to CD80/CD86 on the surface of antigen-presenting cells, preventing CD80/CD86 from binding to CD28 on the surface of T cells. It thereby interferes with the activation of APCs on T cells and blocks pathogen clearance, leading to infection persistence (20, 66).

During the transition from the acute to the chronic phase of osteomyelitis, the function of CD8⁺T cells is impaired in this process and then becomes a participant in the development of chronic osteomyelitis.

It has been found that during chronic osteomyelitis, the killing capacity of cytotoxic T cells was greatly diminished in response to the sustained stimulation of large amounts of antigens and inflammatory factors. These T cells with immunosuppressive properties are called exhausted CD8⁺T cells (66). Exhausted CD8⁺T cells produce fewer cytokines such as IFN-γ, TNF-α, IL-12, IL-18, and IL-12. At the same time, the production of cytotoxic effectors such as granzyme and perforin is also reduced, which causes the inefficient elimination of pathogens (67). Importantly, exhausted CD8⁺T cells express various suppressor receptors such as PD-1, LAG-3, 2B4, CD160, TIM3, and TIGHT (67–69). These suppressor receptors continue to exert immunosuppressive effects in persistent infections. Although the immunosuppressive effect of a single inhibitory receptor is limited, exhausted T cells often express multiple inhibitory receptors, such as PD-1 and LAG-3, which synergistically exert a more potent immunosuppressive effect than a single receptor (70). In addition, the metabolic status of exhausted CD8⁺T cells is altered to adapt to the environment of chronic infection. Normal CD8⁺T cells rely primarily on glycolysis for energy (71, 72); however, exhausted CD8⁺T cells suppress the activation of AKT in the energy synthesis pathway as well as mTOR activity by CTLA-4 or PD-1 (73, 74), resulting in a shift to fatty acid metabolism for energy. It is compatible with the environment of chronic infection that allows the cells to survive in nutrient-deficient environments (75, 76). This suggests that reversing the change in the energy synthesis pathway in CD8⁺T cells or inhibiting fatty acid metabolism in CD8⁺T cells could be a new immunotherapy prospect to reduce the number of exhausted CD8⁺T cells and rescue immunosuppression statutes.

Recently, studies have shown that a difference in transcription factor expression exists between normal effector CD8⁺T cells and exhausted CD8⁺T cells. In recent years, transcription factors are considered to be closely associated with this functional transformation of CD8⁺T cells, including T-bet, EOMES, and TCF1.

T-bet and EOMES are a pair of transcription factors that influence the differentiation fate of initial CD8⁺T cells. When initial CD8⁺T cells express more T-bet than EOMES, they will differentiate toward terminal effector CD8⁺T cells. Conversely, if EOMES are expressed more, initial CD8⁺T cells will differentiate toward memory T cells. The continued high expression of EOMES leads to progressive CD8⁺T cell exhaustion (77). Some researchers believe that a persistent high EOMES expression may be associated with high PD-1 expression, which has been shown in upregulation during chronic osteomyelitis (70, 78).

TCF1 is a T cell-specific transcription factor that is thought to be essential for the differentiation and maintenance of exhausted CD8⁺T. In the early stages of chronic infection, TCF1 is co-expressed with CXCL5 in CD8⁺T cells, then T cells gradually exhibit an exhausted CD8⁺T cell profile. CD8⁺T cells with TCF1 knockdown show a tendency to differentiate toward effector CD8 ⁺ T cells (79, 80). PD-1 expression had a positive effect on TCF1 expression. Thus, the expression of T-bet, EOMES, and TCF-1 on exhausted CD8⁺T cells was all closely associated with PD-1 (81). However, the specific mechanisms by which these transcription factors affect CD8⁺ T cells and the mechanisms by which PD-1 interacts with T-bet, EOMES, and TCF1 remain unclear. It has been reported that CTLA-4 secreted by Treg is able to disrupt the CD80/PD-L1 heterodimer on antigen-presenting cells, resulting in an increase of free PD-L1 (66). Therefore, we speculate that increased PD-L1 activates more PD-L1/PD-1 signaling in osteomyelitis, which may be involved in regulating EOMES expression and mediating the generation of exhausted CD8⁺T cells.

Th17 and Treg cells and Th17/Treg-related immunosuppression are strongly associated with osteomyelitis persistence and bone destruction. Th17 cells, with RORγt on its surface as a unique marker, promote inflammation and inhibit osteoclastogenesis. The Treg surface marker is Foxp3, which is also a transcription factor, and deletion of Foxp3 leads to a reduction in the number of Treg cells. Treg negatively regulates the inflammatory response and also promotes osteoclastogenesis (13, 82–84).

S. aureus is able to activate Treg cells to suppress the host immune response. Sophia Björkander et al. demonstrated that S. aureus activated T cells via superantigens, inducing the large secretion of cytokines in T cells, which activate Treg cells, thus suppressing the immune response (85).

Th17 cells are differentiated from the initial T cells. TNF-α and other inflammatory factors, such as IL-1, IL-6, IL-21, and IL-23, enhance RORγt expression and inhibit Foxp3 expression, which induces the differentiation of initial T cells to Th17 cells (83). Thus, we believe that affecting the T cell differentiation pathway by blocking the secretion of TNF-α, IL-1, and IL-6 can be a strategy to control the inflammation and tissue destruction induced by Th17 cells.

In conclusion, different subtypes of T cells may play different roles in osteomyelitis. CD4⁺T cells interfere with normal immune reaction through the expression of CTLA-4; therefore, how S. aureus affects the T cell expression of CTLA-4 and whether its virulence factors have an impact on CTLA-4 expression in T cells are critical to the study of a new therapeutic target in osteomyelitis. CD8⁺T cells turn into exhausted CD8⁺T cells in osteomyelitis. T-bet, EOMES, and TCF1 are some crucial factors that are associated with an attenuated immune reaction. Treg cells and Th17 cells have the opposite effect, and how to mediate their function in osteomyelitis remains to be explored.

Th17 and Treg cells not only have opposite immune effects in the inflammatory process but also play distinct roles in bone production and destruction.

Treg cells inhibit osteoclast production and thus reduces bone destruction in osteomyelitis, but the mechanism for this remains controversial. Some studies believe that Treg cell inhibits osteoclast production mainly through decreasing the secretion of TGF-β, IL-4, and IL-10 (83). Other studies suggest that Treg cells inhibit osteoclastogenesis primarily through CTLA4-dependent intercellular contacts to reduce bone destruction. Treg-derived CTLA-4 binds to B7-1/B7-2 (CD80/86) on the surface of osteoclast precursors and impairs their differentiation to osteoclasts (86). However, Abdul et al. demonstrated that Treg cells also promoted osteoblastogenesis to accelerate bone remodeling (84). The Wnt signaling pathway is a pathway that regulates osteoblastogenesis. Wnt10b, a ligand for Wnt, is a potent enhancer of the regulation of osteoblast proliferation. Under butyrate conditions produced by Lactobacillus, Treg cells are involved in regulating the activation of the Wnt10b promoter, thus promoting Wnt10b expression (84).

Th17 cells mediate the formation of osteoclasts in various ways. First, the cytokine IL-17 secreted by Th17 cells not only directly increases the expression of RANKL in osteoblasts but also induces the secretion of IL-6, TNF-α, and IL-1, promoting the expression and activation of RANKL in osteoblasts. In addition, Th17 cells can also secrete RANKL (87). In conclusion, osteoblastogenesis is regulated by Th17 cells and promotes bone destruction during inflammation.

Therefore, several cytokines like CTLA-4, IL-17, and IL-6 and signaling pathways like Wnt and RANKL/RANK are involved in the process of Tregs cells and Th17 cells mediating osteoclastogenesis, but other cytokines and pathways in this process are unknown and still remain to be found out. In this review, we figure out the relationship among these factors.

Th17 cells promote inflammatory response and bone destruction, whereas Treg cell inhibits the degree of inflammation and promotes bone formation. Actually, the Th17/Treg ratio is modulated by different conditions. When intense inflammation occurs, inflammatory cells secrete large amounts of inflammatory factors such as IL-6. Inflammatory factors represented by IL-6 stimulate the production of Th17 cells via the STAT3 pathway, whereas in the absence of IL-6, the production of Th17 cell is inhibited (88–90).

Hypoxia-inducible factor α (HIF1α) affects Th17/Treg ratio in a hypoxic environment (91). HIF1α not only increases RORγt expression to promote Th17 cell differentiation but also inhibits Foxp3 expression via the proteasomal degradation pathway under hypoxic conditions, resulting in a relative increase in RORγt expression (92). In summary, IL-6 and HIF1-α, which are produced under hypoxic conditions caused by intense inflammation, regulate Th17/Treg ratio through different pathways, implying that more Th17 cells are induced in an intense inflammation-related hypoxic environment. Conversely, in a more mild, chronic, normoxic inflammatory environment, initial T cells are more inclined to differentiate into Treg cells. Therefore, an ideal environment is critical to balance the Th17/Treg ratio.