Macrophage fate: to kill or not to kill?

ABSTRACT

Macrophages are dynamic innate immune cells that either reside in tissue, serving as sentinels, or recruited as monocytes from bone marrow into inflamed and infected tissue. In response to cues in the tissue microenvironment (TME), macrophages polarize on a continuum toward M1 or M2 with diverse roles in progression and resolution of disease. M1-like macrophages exhibit proinflammatory functions with antimicrobial and anti-tumorigenic activities, while M2-like macrophages have anti-inflammatory functions that generally resolve inflammatory responses and orchestrate a tissue healing process. Given these opposite phenotypes, proper spatiotemporal coordination of macrophage polarization in response to cues within the TME is critical to effectively resolve infectious disease and regulate wound healing. However, if this spatiotemporal coordination becomes disrupted due to persistent infection or dysregulated coagulation, macrophages’ inappropriate response to these cues will result in the development of diseases with clinically unfavorable outcomes. Since plasticity and heterogeneity are hallmarks of macrophages, they are attractive targets for therapies to reprogram toward specific phenotypes that could resolve disease and favor clinical prognosis. In this review, we discuss how basic science studies have elucidated macrophage polarization mechanisms in TMEs during infections and inflammation, particularly coagulation. Therefore, understanding the dynamics of macrophage polarization within TMEs in diseases is important in further development of targeted therapies.

INTRODUCTION

Macrophages are among the first immune cells to respond to foreign invaders, damaged tissues, or tumor cells (1–3). Macrophages are responsible for phagocytizing and killing pathogens, clearing dead cells, and healing damaged tissue (1–4).

Based on their origin, macrophages are differentiated into tissue-resident macrophages (TRMs) or monocyte-derived macrophages (MDMs). TRMs originate during embryonic development and proliferate in the tissue (Fig. 1.1) (5, 6). On the other hand, MDMs are generated from peripheral blood monocytes which are recruited into damaged and/or infected tissue (Fig. 1.2 and 1.3) (7, 8). Macrophages exhibit plasticity by polarizing into a continuum of pro- (M1-like) and anti-inflammatory (M2-like) phenotypes upon exposure to microenvironmental cues (Fig. 1.4) (9, 10). Macrophages display receptors that recognize these cues in the TME, which ultimately induce polarization and modulate M1-like microbicidal and M2-like repair responses (Fig. 1.4) (9, 11–16).

Fig 1.

Tissue microenvironment signals induce M1 and M2 macrophage polarization on a spectrum. (Step 1) In the initial response to infection and/or damage, tissue-resident macrophages (TRMɸs) serve as sentinels by sensing pathogen-associated molecular patterns (PAMPs) and damage-associated molecular patterns (DAMPs) to stimulate CCL2 for monocyte recruitment. (Step 2) Monocytes extravasate into the tissue where (Step 3) Macrophage-colony stimulating factor (M-CSF) differentiates monocytes into naive (M0) monocyte-derived macrophages (MDMɸs). (Step 4) Factors including PAMPs, DAMPs, and cytokines produced by T_H1 and T_H2 cells then induce M1 and M2 polarization on a spectrum. IFN-γ, LPS, and GM-CSF induce M1-like polarization leading to a signaling cascade that enables transcription factors STAT1, activator protein-1 (AP-1), NFκB, IRF3, and IRF5 to activate proinflammatory genes. Cytokines and chemokines produced by M1s activate CD8⁺ cytotoxic T cells (T_c), T_H1, natural killer (NK) cells, and neutrophils (PMN). Virus-infected macrophages can also present viral antigens on their major histocompatibility complex-I (MHC-I) to T_C cells, resulting in granzyme release that induces apoptosis. Macrophages infected with bacteria or fungi can present their antigen on MHC-II to T_H1 cells resulting in IFN-γ secretion. IFN-γ promotes phagocytosis and upregulates iNOS to produce NOs degrading the pathogen. IL-4, IL-13, immune complexes, and glucocorticoids induce M2-like polarization. For M2-like polarization, signaling cascades activate transcription factors IRF4, STAT3, STAT6, PPARɑ, and CCAAT/enhancer-binding protein beta (C/EBPβ) to induce anti-inflammatory genes. Cytokines and chemokines produced by M2s dampen M1s, NKs, and PMNs inflammatory response. Furthermore, these factors activate T_H2 and regulatory T cells (T_reg) to secrete cytokines to suppress phagocytosis, ultimately resulting in pathogen survival and persistence. (Step 5) At the same time, endothelial cells produce IL-6, which stimulates hepatocytes to produce Fg and infiltrate into the tissue to form fibrin clots by proteolytic activity of thrombin for tissue repair.

M1-like macrophages generally secrete proinflammatory cytokines, produce high amounts of reactive nitrogen and oxygen species for antimicrobial and tumoricidal responses, activate inflammatory T-helper-1 (T_H1), and promote robust phagocytic ability (Fig. 1.4). By contrast, M2-like generally resolve inflammatory responses, by phagocytizing apoptotic cells, promoting tissue repair and remodeling, as well as inducing anti-parasitic responses (Fig. 1.4) (4, 9, 15, 17, 18). Based on changes in the TME, macrophage polarization is a highly dynamic process and can switch between different phenotypes beyond the M1-M2 dichotomy (10, 19). With opposing activities of killing and repairing, it is critical to tightly regulate M1-M2 polarization to prevent disease since dysfunctional M1-M2 polarization is associated with adverse pathologies (20). For example, sustained M1-like polarization is implicated in inflammatory diseases such as acute rheumatoid arthritis, systemic lupus erythematosus, inflammatory bowel diseases, atherosclerosis, and even exacerbating SARS-CoV-2 infection (COVID-19) (21–23). Furthermore, excessive M2-like polarization is associated with cancers and asthma and promotes bacterial and fungal infections (24, 25). Interestingly, macrophage response plays a critical role in responding to vascular and tissue injury by interacting with coagulation factors to carefully orchestrate inflammation and wound healing (26). Thus, macrophage dysfunction can contribute to promoting coagulopathy-associated diseases, especially hemophilic and thrombotic diseases (26). Therefore, there is a need to better understand how macrophage polarization is modulated by coagulation factors in both wound healing and coagulopathies, which this review will discuss.

Taking advantage of macrophage plasticity, therapies have aimed at repolarizing macrophages between M1s and M2s to generate immunotherapeutic strategies to improve clinical outcomes in favor of the patients (27). However, to optimize this therapeutic approach, it is crucial to understand the mechanisms of how the tissue microenvironment (TME) during diseases drives macrophage polarization toward clinically unfavorable outcomes. In this review, we will discuss how infections and coagulation modulate macrophage polarization in the TME. Thus, this review will provide a better understanding of the dynamics of macrophage polarization within TMEs in diseases, which is important in the future development of therapies.

INDUCTION OF MACROPHAGE POLARIZATION

TRMs are immune sentinels that detect pathogen-associated (PAMPs) and damage-associated (DAMPs) molecular patterns. Upon recognition of PAMPs and DAMPs, TRMs produce cytokines and chemokines such as CXCL1 and CXCL2 for neutrophil recruitment and CCL2 for monocyte infiltration to damaged/infected tissue (Fig. 1.1) (15, 28–30). During acute inflammation, monocytes are recruited from the bone marrow into affected tissues (Fig. 1.2), which then differentiate into MDMs upon exposure to macrophage-colony stimulating factor (M-CSF/CSF-1) (Fig. 1.3) (31). In this altered TME, both TRMs and MDMs are exposed to cues that then spatiotemporally polarize these macrophages to a spectrum of phenotypes tailored to specific functions (Fig. 1.4). Upon receiving these cues, intracellular signaling cascades will alter the transcriptional programming, promoting specific metabolic and enzymatic pathways needed for M1-like proinflammatory or M2-like anti-inflammatory response. To understand this polarization, we discuss how the spectrum of phenotypes is induced and regulated and the challenges of describing macrophage polarization.

Induction of M1 macrophages

M1-like macrophages generally have a proinflammatory activation pattern where they (i) secrete inflammatory cytokines and chemokines; (ii) produce reactive oxygen (ROS) and nitrogen species (RNS); and (iii) promote antigen presentation to activate T_H1 response for anti-tumorigenic and antimicrobial activity (9, 10, 15). Depending on the stimulus in the TME, macrophages can polarize toward M1s displaying a spectrum of proinflammatory phenotypes (32). Discoveries made from in vitro studies using human and murine macrophage-like cell lines and ex vivo studies using bone marrow-derived macrophages (BMDMs), murine peritoneal macrophages, and human macrophages from peripheral blood mononuclear cells led to the realization that different inducers can polarize macrophages on a spectrum of M1-like phenotypes (32). Each activating inducer and its corresponding murine and human markers are described in specific detail in the nomenclature and experimental guidelines by Murray et al (32).

It has been established that stimulation of cultured murine and human peripheral blood macrophages with interferon-γ (IFN-γ) exert a clear-cut M1 polarization mediated by signal transducers and activators of transcription 1 (STAT1) and interferon regulatory factors (IRF)-1, -5, and -8 signaling (32–34). Specifically in murine macrophages, IFN-γ binding to IFNGR receptor activates JAK1/JAK2 to promote STAT1 binding to gamma interferon activation site (GAS) elements in the promoter of the Nos2 gene (35–37). This pathway leads to M1s expressing a murine-specific marker inducible nitric oxide synthase (iNOS), which metabolizes macrophages’ arginine to nitric oxides (NOs) to kill the phagocytized pathogen (35–37). In addition to NO production, all macrophages express the NADPH oxidase complex (NOX2), which catalyzes superoxide production upon IFN-γ activation (38, 39). The generation of these superoxides is important for producing secondary products, such as hydrogen peroxide and hydroxyl radicals, which possess stronger bactericidal effects (38). For this to occur, activation of NOX2 requires translocation of subunits p40^phox, p47^phox, p67^phox, and Rac from the cytosol to the integral membrane heterodimer Flavocytochrome b₅₅₈ (38). Flavocytochrome b₅₅₈ consists of two membrane subunits gp91^phox and p22^phox that mediates electron transfer from NADPH to molecular oxygen to produce superoxides (38). In murine and human macrophages, IFN-γ promotes trafficking of gp91^phox and p22^phox to the plasma membrane through the endocytic pathway resulting in increased superoxide production (39). Aside from iNOS and NOX activation, IFN-γ activates STAT1 and NF-κB in separate pathways to also produce inflammatory cytokines and chemokines including CXCL9-11, CCL5, interleukin (IL)−1β, IL-6, IL-12, IL-18, IL-23, and TNF-ɑ (9, 10, 15).

M1s are activated by the interaction of the toll-like receptor (TLR)-4 and co-receptor CD14 with lipopolysaccharide (LPS; Gram-negative bacterial cell wall component), which then induces nuclear factor-kappa B (NF-κB), activating protein-1 (AP-1), and interferon regulatory factor (IRF)−3 (Fig. 1.4) (10). Furthermore, TLR-4 signaling can also induce IRF3, which then releases IFN-ɑ and -β (40–43). Much like exposure to IFN-γ, stimulation with LPS alone or in combination with IFN-γ will also produce these proinflammatory cytokines and chemokines (15). Common with both inducers is NF-κB activation, which is a key transcription factor of M1 polarization and induces TNF-ɑ, IL-1β, IL-6, IL-12, cyclooxygenase-2 (COX-2), hypoxia-inducible factor 1ɑ (HIF-1ɑ), and suppressor of cytokine signaling 3 (SOCS3) (41, 42). These secreted or expressed factors are known to inhibit production of cytokines, chemokines, and transcription factors that promote M2 polarization (9, 10).

Also, granulocyte macrophage-colony stimulating factor (GM-CSF), produced by fibroblasts, endothelial cells, recruited innate immune cells, and T cells, can shift macrophages toward the M1-like phenotype as inflammation or infection progresses (31). In response to infection or tissue damage, both M-CSF and GM-CSF induce macrophage differentiation (31). During homeostasis, M-CSF is ubiquitously produced by tissue-resident cells to differentiate newly recruited monocytes into macrophages upon initial infection and tissue injury (31). As inflammatory conditions prolong, GM-CSF levels then quickly increases to stimulate recruitment, survival, and activation of myeloid cells, particularly M1 polarization (31, 44). GM-CSF is recognized by macrophages’ GM-CSF receptor (CSF-2R), which activates IRF5 to drive the secretion of IL-12 and IL-23, promoting inflammatory T_H1 and T_H17 cell response (45). In an inflamed TME, GM-CSF plays a role in regulating macrophage antimicrobial defense and wound healing (31, 46–48). For example, in an inflammatory bowel disease mouse model, it was shown that the group 3 innate lymphoid cells were the major producers of GM-CSF, leading to induce M1 polarization and bactericidal response against Citrobacter rodentium while suppressing M2 pro-fibrotic response (48).

Induction of M2 macrophages

In contrast to M1s, M2-like macrophages are critical for resolving inflammation by (i) suppressing proinflammatory response; (ii) activating anti-inflammatory T_H2 cells; (iii) clearing apoptotic cells from damaged tissue; and (iv) repairing and remodeling tissue to restore homeostasis (49). Moreover, based on stimuli in the TME, macrophages can polarize to diverse M2-like phenotypes with different functions in suppressing proinflammatory response to promote wound healing (Fig. 1.4) (49, 50).

Much like how IFN-γ induces M1s, IL-4 and/or IL-13 exert an antagonistic effect on IFN-γ to promote M2 polarization mediated by STAT6 signaling (32, 51). STAT6 activation, in turn, then induces the expression of CD206 (Mannose receptor C-type 1, Mrc1) and CD163 (scavenger receptor for hemoglobin and haptoglobin) on both human and mouse macrophages (10, 52). Furthermore, mouse genetic studies investigating M2 polarization found expression of murine-specific M2 markers such as Chitinase 3-like 3 (Chi3l3, Ym1; also known as eosinophil chemotactic factor-lymphocyte), Arginase-1 (Arg1), and Resistin-like ɑ (Retnla, Fizz1) (25, 53, 54). In addition, STAT6 cooperates with transcriptional factor KLF4 to antagonize the M1-mediated STAT1 signaling pathway, thereby activating anti-inflammatory genes and leading to the production of IL-10, transforming growth factor (TGF)-β, CCL17-18, CCL22, and CCL24 (10, 15, 52). This activation leads macrophages to have a wound-healing phenotype with the ability to promote endocytic activity, cell growth, and tissue repair by secreting collagen precursors, stimulating fibroblasts, and secreting platelet-derived growth factor (PDGF) (49, 50). They also produce high levels of Arginase-1 (Arg1), competing with iNOS for arginine to produce polyamines needed for tissue repair (49).

Interestingly, M2-like polarization can also be induced by a combination of immune complexes with LPS, TLR, and IL-1R agonists leading to a subset that expresses both pro- and anti-inflammatory markers, such as iNOS, Arg1, CD86, CD68, MHCII, and CD163 (52, 55). In addition, they also produce anti-inflammatory IL-10 and CCL1 to suppress inflammation by promoting T_H2 and T_reg cell responses (52). M2s induced by immune complexes are found to have immunoregulatory effects with protective and pathogenic roles depending on TME (52, 55). For example, a study showed that macrophages induced with both LPS and IgG antibodies resulted in M2 macrophages with pro- and anti-inflammatory profiles (56, 57). These specific M2-like macrophages were shown to have a role in preventing excessive fibrosis ex vivo and in vivo, thereby reducing tissue injury in a myocardial ischemia in vivo model (56, 57).

M2-like macrophages can also display a pro-resolving phenotype induced by glucocorticoids (GCs) (52). GCs, including cortisol, dexamethasone, and prednisone, have a potent ability to suppress proinflammatory responses through nuclear glucocorticoid receptors (GR) in macrophages (58). GCs, with their lipophilic structure, penetrate the plasma membrane and bind to the cytosolic GR forming a GC/GR complex that will then translocate into the nucleus (59). This complex will then bind to GC response elements (GREs) in the promoter regions of GC-associated genes that promote anti-inflammatory response (59). To inhibit pro-inflammatory genes, the GC/GR complex can either bind to “negative” GREs upstream of pro-inflammatory genes or directly interact (or tether) with transcription factors to prevent activation (59). Through this mechanism, GCs inhibit transcription of M1-producing cytokine genes Il1b, Il6, Il12, and tnf and downregulate iNOS and COX-2 in both human and mouse macrophages (59, 60). GCs are found to induce M2 polarization similar to macrophages treated with IL-4 and IL-13, which enable IL-10 and TGF-β production and upregulates both mannose and CD163 receptor (59). Together, this allows these macrophages to resolve inflammation by promoting phagocytosis of apoptotic cells during tissue repair (59).

In summary, it is critical to understand the TMEs and their diverse signals to further dissect macrophage polarization, function, and their role in disease outcome, since dysregulated polarization may result in exacerbated pathologies.

Challenges to describing macrophage polarization

Recently, researchers in the macrophage biology field acknowledged the challenges to describing macrophage polarization due to the lack of consensus on standardizing how to define polarization states and their resulting phenotypes in vitro, ex vivo, and in vivo experiments (32). First, the widespread use of the M1/M2 dichotomy or “classical”/“alternative” activation over decades and to this day pushes the oversimplification that macrophage polarization occurs on these two extreme ends of the spectrum, where in reality it does not. “Classical” and “alternative” activation terms first encompassed from in vitro studies that demonstrated IFN-γ and IL-4 differentially polarize macrophages to pro- or anti-inflammatory phenotypes, respectively (61, 62). Several years later, it was discovered that peritoneal macrophages isolated from C57BL/6 mice (mice with elevated proinflammatory T-helper-1 (T_H1) response) produced higher bactericidal NOs when stimulated with either IFN-γ or LPS in vitro (63). By contrast, macrophages isolated from BALB/C mice (mice with higher anti-inflammatory T_H2 response) yielded low amounts of NOs when stimulated with IFN-γ and/or LPS (63). Importantly, it was found that murine peritoneal macrophages from these two different strains showed opposite metabolic pathways on arginine: where macrophages from C57BL/6 mice metabolize arginine to NOs via iNOS, while macrophages from BALB/C mice metabolize arginine to ornithine via Arg1. Due to these observations, the M1-M2 dichotomy was initially proposed, which mirrored the T_H1/T_H2 responses in that T_H1 cells’ production of IFN-γ promotes M1s, while T_H2 cells produce IL-4 to polarize M2s (63). Although now it is known that sources of IFN-γ and IL-4 can come from other immune cells, especially innate lymphoid cells (64).

Further advancement through in vitro studies on murine and human macrophages with different stimuli revealed diverse phenotypes that led to the development of subsets, such as M2a, M2b, M2c, and M2d to name a few examples (52, 65). However, this nomenclature has generated inconsistencies and confusion on which cell surface, intracellular, and secreted marker bins macrophages into these subsets. In addition, it is difficult to determine whether macrophages binned into these subsets are terminally differentiated into these subsets and whether they are indeed functionally distinct from each other. To address the pitfalls in describing macrophage polarization, a group of experts in the macrophage biology field met at the International Congress of Immunology in 2013 and provided nomenclature and experimental guidelines, which are summarized in reference (32).

Although in vitro studies have contributed to defining M1 and M2 phenotypes, the major challenge is that activation signatures defined in vitro can be modulated by mixed signals that are overlooked in vivo. Therefore, translating in vitro results to in vivo mouse and human disease models is problematic due to the complexity of factors in the TME that polarize macrophages beyond the M1 and M2 extremes. Considering that macrophages display plasticity, changes in the TME will cause macrophages to transition into intermediate states between M1s and M2s. Thus, the limitation of the simple bipolar M1-M2 terminology is that these two states do not exist alone in TMEs. Hence, in this review, we describe M1s and M2s as general activation patterns and acknowledge that a continuum of phenotypes exists in vivo and in human disease studies.

MACROPHAGE POLARIZATION DURING INFECTIONS

During infections, macrophages enhance their antimicrobial responses using their repertoire of pattern recognition receptors (PRRs) to engage with pathogens directly, or indirectly with opsonins, inducing M1 polarization and phagocytosis (Fig. 1.4) (66). Macrophages then amplify their antimicrobial response by producing highly reactive ROS and RNS, activating T_H1 and T_H17 cells via antigen presentation, and secreting proinflammatory cytokines to promote immune cell recruitment (15, 24, 67). Due to M1’s antimicrobial responses, some pathogens have developed immune evasion mechanisms that manipulate macrophage polarization and function, thereby preventing phagocytosis and killing to promote colonization and infection (25).

Viral infections

Viruses also take advantage of M1-M2 polarization to promote their virulence and infection (Fig. 2). When macrophages encounter viruses, macrophages polarize to M1s, which leads to activating IRFs, producing type I interferons (IFN-ɑ and -β), promoting natural killer cells (NK) and T cell activation (10, 68). In addition, M1s utilize intracellular RIG-I-Like receptors (RLRs) to recognize viral RNA, leading to the induction of IFNs and IFN-stimulated genes (69). For example, in a mouse West Nile Virus infection model, upon RLRs recognition of viral RNAs, M1 programming is activated to control the infection (70).

Fig 2.

Pathogens can modulate macrophage polarization leading to microbial clearance or survival. Bacteria, viruses, and fungi species can promote macrophages toward (A) M1 or (B) M2 polarization on a continuum affecting the outcome of infection.

Although M1s are essential in acute infection, excessive M1 function by prolonged inflammatory cytokine secretion results in immune hyperactivation, tissue damage, organ failure, viral persistence, and death (71). During the COVID-19 pandemic, in most cases of SARS-CoV-2 infections, patients were able to clear the infection, which correlated with the presence of IFNs, IL-1β, IL-6, and TNF-ɑ (23, 72). Contrarily, hospitalized patients with excessively high levels of IL-6 and TNF-ɑ in serum correlated with severe disease, including complications of pneumonia and death (72). It has been speculated that the hyperinflammatory signature of the airway myeloid cells promotes an excessive M1 function, which perpetuates immune cell recruitment and lung inflammation, leading to severe disease (73). Since various immune cells produce these cytokines, it is difficult to discern if M1s are solely responsible for manifesting cytokine release syndrome (termed as “cytokine storms”).

Viruses can also modulate multiple macrophage polarization states to promote viral dissemination. Viruses such as Hepatitis B, Hepatitis C, and Epstein-Barr Virus (EBV) can induce M2 polarization (Fig. 2B) (74–77). Human cytomegalovirus (HCMVs) is able to establish latency after initial infection (78). HCMVs target monocytes for viral entry and promote monocyte survival by differentiating them into hybrid M1/M2 macrophages (78, 79). This atypical M1 polarization is characterized by expressing an M1 transcriptional profile but also upregulating anti-inflammatory factors such as IL-1R antagonists and IL-10. This hybrid phenotype results in pathogen clearance reduction, promoting viral entry, intracellular survival, and transendothelial migration (79). Another study further confirmed that different HCMV strains induced a PI3K-SHIP1-Akt signaling pathway in monocytes, resulting in a hybrid M1/M2 phenotype (80).

A way for the host to orchestrate M1/M2 response during viral infection is by sending signals that reduce M1 inflammatory cytokine secretion while promoting M2 functions in repairing damaged tissue to restore homeostasis. During the repair phase, apoptotic cell clearance (efferocytosis) by macrophages activates anti-inflammatory programming, which is triggered by degradation products of apoptotic cells (ACs) within phagolysosomes (81). This orchestration into M2s is critical to prevent inflammation-induced tissue damage. Interestingly, a study found that M2s that engulfed SARS-CoV-2-infected-ACs reduced expression of genes associated with tissue remodeling and immunoregulatory functions including ccl18, cd206, mmp9, pparg, and cd163, but increased expression and robust secretion of IL-6 and IL-1β, which promotes inflammation (81). Of importance, CD206 (Mrc1), a mannose receptor on M2s, was decreased on these specific macrophages and impaired efferocytosis, which may provide an explanation to viral persistence in severe COVID-19 (81).

Moreover, the presence of anti-inflammatory mediators during acute viral infection renders the host susceptible to developing secondary bacterial infections (82). Recently, a study identified that excessive induction of the anti-inflammatory CYP450 lipid metabolite activates the peroxisome proliferator-activated receptor (PPAR) via transcription factor PPARɑ (83, 84). This signaling dampened TLR stimulation, leading to suppression of NFκB-mediated response in a mouse model of secondary S. aureus lung infection post-influenza (83, 84). Furthermore, PPARɑ activation promoted M2 polarization and correlated with inhibiting S. aureus clearance (84). These findings demonstrate that viruses actively capitalize on both M1s and M2s to promote persistent infection.

Bacterial infections

M1 targets pathogens during acute phase of infection by phagocytosing and degrading bacteria through ROS and RNS within phagosomes (Fig. 1.4). Specific pathogens are known to induce M1 polarization including E. coli, Salmonella typhi, Helicobacter pylori, Enterococcus faecalis, Staphylococcus aureus, Listeria monocytogenes, Mycobacterium tuberculosis, and Mycobacterium ulcerans (Fig. 2A) (85–91).

LPS is among the virulence factors that influence macrophage polarization, which is a major component of the outer membrane of gram-negative bacteria that is recognized by TLR-4 and its coreceptor CD14 and MD-2 (92). Pathogens such as E. coli, S. typhi, and H. pylori possess LPS, which activates NF-κB, induces proinflammatory cytokine production, promotes phagocytosis, and enhances intracellular pathogen degradation through TLR-4/CD14 signaling (88, 93, 94). In an uncomplicated urinary tract infection mouse model, LPS from uropathogenic E. coli (UPEC) upregulates expression of CD14 for binding on macrophages to induce antibacterial responses via NFκB and p38 mitogen-activated protein kinase resulting in decreased UPEC burden in bladders (95). In the same study, UPEC infection was exacerbated in CD14-deficient mice bladders, which was associated with a decrease in CD14-dependent pathways including immune cell trafficking, proinflammatory cytokine production in macrophages, and IL-17 signaling (95). Thus, CD14 as a TLR-4 coreceptor is pivotal in amplifying M1-like macrophage response.

In an active infection in the TME, invasive pathogens have distinct ways to modify the host extracellular matrix (ECM) by (i) producing tissue-degrading enzymes called invasins; (ii) hijacking host proteolytic systems; or (iii) capitalizing on host immune cells’ production of proteases to facilitate migration and colonization throughout the tissue (96, 97). This ECM degradation enables macrophages to both recognize pathogens through its PRRs and sense physical alterations in the ECM through their mechanosensitive ion channels (MSICs) (96–99). Interestingly, a mechanosensitive ion channel, Piezo1, expressed on mouse macrophages, senses these alterations in the ECM stiffness, which leads to the increasing influx of Ca²⁺ into macrophages to promote M1-like responses, particularly inducing IFN-γ production and iNOS upregulation (98). Another group demonstrated that E. coli infection-induced TLR-4 signaling increased Piezo1 activity in mouse macrophages ex vivo and in vivo (100). LPS stimulation by E. coli infection triggered the assembly of TLR-4 and Piezo1 as a coreceptor to induce Ca²⁺ influx leading to cytoskeleton remodeling for phagocytosis, ROS production, and M1-like bactericidal response through the CaMKII-Mst1/2-Rac1 signaling axis (100). Deficiency in Piezo1 promotes wound healing M2-like macrophages (98). This suggests that MSICs during bacterial infection play a role in coordinating with TLRs in modulating macrophage polarization; however, this needs further evaluation.

Although several gram-negative bacterial species have LPS and other virulence features that generally promote M1 polarization through TLRs signaling, some species, particularly H. pylori and Francisella tularensis, can evade the innate immune response by displaying structural variations of LPS and flagellin that impede recognition by TLR-2, -4, and -5 on macrophages (101–103). The structure of LPS consists of a polysaccharide O-antigen chain, a core oligosaccharide, and lipid A, which serves to anchor LPS to the plasma membrane (104). Unlike E. coli LPS structure, the biosynthesis of H. pylori O-antigen leads to a unique variation called the Lewis system antigens, which has been shown to subvert host immune response (101, 105). Thus, it is this alteration in the O-antigen that causes H. pylori LPS to antagonize TLR-4 preventing NF-κB activation in human macrophages (106, 107). Although this pathogen was able to be phagocytosed by macrophages, H. pylori is still highly resistant to killing within phagosomes by preventing phagosome maturation, utilizing bacterial Arg1 to outcompete macrophages’ iNOS for arginine, and inducing mitochondrial-associated apoptosis (108–111).

Another evasion strategy that bacteria utilize is to drive macrophages toward M2 polarization (Fig. 2B) (112). For example, S. enterica serovar Typhimurium, a foodborne pathogen, produces the effector protein, SteE, to induce M2 polarization (112, 113). SteE activates STAT3, which increases IL-10 levels resulting in bacterial intracellular replication in vitro and increased bacterial burden in a murine model (112, 113). Another pathogen, F. tularensis, a causative agent of the zoonotic disease tularemia, promotes M2 polarization through TLR-2 signaling to induce Arg1, allowing bacterial replication inside macrophages (114). This specific pathogen has an atypical LPS structure that renders it inert instead of serving as a potent endotoxin due to its inability to interact with host TLR-4 and its coreceptors CD14 and MD-2 (103, 115). Despite failure to induce TLR-4 signaling, F. tularensis was still able to be phagocytosed by macrophages’ mannose receptors which then formed phagosomes containing the pathogen (116). In addition, once inside the cell, F. tularensis was able to upregulate microRNA-155 to downregulate TLR adapter protein MyD88 and phosphatase SHIP-1, consequently suppressing downstream proinflammatory response in human macrophages (117). In addition, this pathogen expresses acid phosphatases that appear to enhance intracellular survival by selectively dephosphorylating the p40^phox and p47^phox subunits of the NADPH oxidase complex, inhibiting ROS production (118). Furthermore, F. tularensis possess the ability to selectively destabilize host mRNA encoding proinflammatory IL-1β, IL-6, and CXCL1 once the pathogen escaped from phagosomes into the cytosol of infected macrophages (119).

Fungal infections

During fungal infections, macrophages display various PRRs that are capable of recognizing β-glucans, zymosans, chitin, and ɑ-mannans on fungal cell walls to induce fungicidal response (120). M1s utilize C-type lectin receptors (CLRs), such as Dectin-1 and Dectin-2, to bind to β-glucans of fungal pathogens and induce ROS production (121, 122). In addition, recognition of fungal cell wall components by cell surface receptors such as TLR-2, TLR-4, TLR-5, Dectin-1, Dectin-2, and Mincle results in the induction of downstream activation of caspase recruitment domain-containing protein 9 (CARD9) (123, 124), which is essential in M1 activation and for induction of vaccine-mediated immunity against Cryptococcus neoformans, a causative agent of pulmonary cryptococcosis (125). Furthermore, CARD9 is also involved in activating NF-κB downstream of Retinoic acid-inducible gene I receptor (RIG-I) (126). In animal studies, CARD9 deficiency skewed macrophages to M2s, leading to decreased NO production and resulted in inability to control replication and dissemination even after vaccination against C. neoformans (125).

PPRs collaboration is critical for triggering macrophage’s inflammatory response against fungi. In the case of Histoplasma capsulatum, this fungus is recognized by both macrophage complement receptor-3 (CR3) and Dectin-1 (127). This interaction induces macrophage secretion of TNF and IL-6, facilitating host responses against disseminated histoplasmosis (127). During Candida albicans infection, CLRs Dectin-1, -2, and Mincle collaborate and activate cytokine production by inflammatory monocytes (128).

As an evasion mechanism upon phagocytosis, fungal pathogens such as C. neoformans, Talaromyces marneffei, and Candida spp. can rewire macrophages to skew toward M2s for their survival and dissemination (Fig. 2B) (129–133). For instance, an ex vivo and in vivo study found that C. neoformans enhanced its virulence by secreting CPL1 and binding to TLR-4 (132). This interaction promoted M2 polarization by propagating IL-4 signaling, which upregulated Arg1 allowing C. neoformans to replicate within macrophages (132). Furthermore, an animal study showed that oral antibiotic treatment induced overgrowth of gut commensal Candida species (129). Interestingly, the fungal overgrowth in the gut had systemic effects, resulting in increasing PGE₂ plasma concentrations, which then polarized alveolar macrophages to M2s and exacerbated allergen-induced airway inflammation (129). Some reports have shown that fungi can produce PGE₂ (134, 135); however, this report does not assess whether the increased levels of PGE₂ is due to fungal production (134, 135) or host cells’ PGE₂ production in response to the Candida overgrowth (136).

Reprogramming of metabolic pathways during macrophage polarization is key to address the energetic and anabolic needs to respond to infections. To support M1s microbicidal functions, M1s elevates glycolytic metabolism by increasing pentose phosphate pathway activity that results in the production of ROS for superoxide burst (130). It was found that the melanin on the surface of Aspergillus fumigatus spores activates HIF1ɑ-dependent glycolysis in macrophages, resulting in increased phagocytosis and pathogen clearance (137).

THE ROLE OF HEMOSTASIS AND COAGULATION IN MODULATING MACROPHAGE POLARIZATION

Role of hemostasis in tissue injury and repair

Physical trauma and infection can cause tissue injury that then initiates hemostasis, which is a mechanism that leads to the cessation of excessive bleeding from damaged blood vessels (138). Hemostasis is a complex regulated process that employs platelet adhesion and production of pro- and anti-coagulant factors, fibrinolytic proteins, chemokines, cytokines, and growth factors necessary for arresting hemorrhage and promoting tissue repair (138, 139). This process is divided into four stages: (i) vasoconstriction; (ii) platelet adhesion and aggregation; (iii) coagulation cascade activation leading to fibrin clotting (thrombosis) (Fig. 3A through C); and (iv) fibrinolysis resulting in fibrin clot dissolution (Fig. 3D) (139). Immediately after initial injury to blood vessels, the damaged vascular endothelium secretes endothelin to trigger vasoconstriction (139). At this stage, activated endothelial cells produce chemoattractants, cytokines, and adhesion molecules, such as collagen and von Willebrand Factor, that augment recruitment, adhesion, and aggregation of platelets, neutrophils, monocytes, and macrophages at the injury site (139). Mediators secreted by platelets from the damaged vessel initiate activation of the coagulation cascade that ultimately lead to fibrin formation that serve as a temporary plug for wound healing (140). To get this fibrin formation, hepatocytes in the liver first produce fibrinogen (Fg) as one of the prominent acute phase proteins that circulate in the blood to the damaged site (Fig. 1.5) (141). Through multiple triggers at these sites, a complex cascade of reactions containing coagulation factors becomes activated through the intrinsic pathway (by contact activation from damaged blood vessels) and the extrinsic pathway (induced by tissue damage) (Fig. 3A and B), which ultimately promotes proteolytic activity of thrombin to polymerize Fg into fibrin clots (Fig. 3C) (142).

Fig 3.

Vascular and tissue injury activates the coagulation cascade leading to fibrin clot formation. (A) Vascular damage activates the intrinsic pathway of the coagulation cascade by proteolytic cleavage of FXII, FXI, and FIX. Activation of FVIII produces FVIIIa to form a complex with FIXa (FIXa:FVIIIa) to subsequently cleave FX into FXa. (B) Tissue injury initiates the extrinsic pathway of the coagulation cascade by activating FIII (Tissue Factor) and FVII to generate the FIIIa:FVIIa complex. The FIIIa:FVIIa complex cleaves FX to produce FXa. (C) Both FXa and FVa activate FII (prothrombin) into FIIa (thrombin). Thrombin proteolytically cleaves soluble Fg fibrinopeptides to form fibrin fibers. FXIIIa crosslinks fibrin fibers to stabilize fibrin clots. (D) For fibrinolysis to occur, urokinase- (uPA) and tissue-plasminogen activator (tPA) activate plasminogen (PG) into plasmin to degrade fibrin clots generating fibrin degradation products. To halt the fibrinolytic process, uPA and tPA are inhibited by plasminogen activator inhibitor-I (PAI-I).

The destruction of tissue leads to alterations in the endothelial cells (ECs) and the ECM that signals monocytes and macrophages to be recruited and adhere to these injury sites (143). At these sites, macrophages serve as key regulators in the tissue healing process by properly amplifying and dampening inflammation, tissue repair/remodeling, removing apoptotic cells, and restoring tissue homeostasis (49, 144, 145). Upon initial exposure to IL-1β, IL-6, IL-8, and TNF-ɑ released by platelets, macrophages upregulate expression of tissue factor (TF) on its cell surface, which is critical in initiating the coagulation cascade (146). At the same time, macrophages itself release proinflammatory cytokines, such as IL-1, IL-6, and TNF-ɑ, to increase immune cell recruitment, upregulate adhesion molecules (i.e., selectins and integrins) to interact with ECs for transmigration into damaged tissue and promote procoagulant factors (26). Subsequently, thrombin released by platelets cleaves G-protein-coupled protease-activated receptors (PARs)-1, -3, and -4 on macrophages to drive proinflammatory M1-like responses (26). Once fibrin clots are formed, macrophages produce tissue factor pathway inhibitors to suppress TF-induced coagulation and eventually clear procoagulant factors to prevent excessive clot formation (26, 147). Since diverse functions are driven by macrophage polarization, dysregulation of these functions modulated by coagulation factors can lead to hemophilic or thrombotic diseases (26). Therefore, it is critical that macrophage function is regulated to prevent the development of these diseases. Due of the complexity of the coagulation cascade, this review will highlight a few of the selected coagulation factors and regulators that been shown to be involved in macrophage polarization.

Tissue injury-induced coagulation activation modulates macrophage polarization for tissue repair

The coagulation cascade is directly involved in modulating macrophage response to orchestrate inflammatory responses for wound healing (148). Macrophages express receptors, such as protease-activated receptors (PARs), TLRs, and integrins, that recognize different coagulation factors (148). Initial vascular injury induces platelets to release thrombin that activates PAR-1 expressed on macrophages resulting in the production of proinflammatory cytokines (148). In this intrinsic pathway, activated FXII produced by damaged vessels appears to have a role in inducing M1 polarization through the urokinase plasminogen activator receptor (uPAR) to secrete TNF-ɑ, IL-1β, IL-6, and IL-6 (149). Further down the cascade, FVIII plays an important role in macrophage differentiation and polarization (150, 151). Deficiency in FVIII or FIX led to prolonged bleeding in patients with hemophilia due to the inability to clot (152). Interestingly, it was observed that the majority of monocytes from the peripheral blood of hemophilia patients failed to differentiate into macrophages upon exposure to either M-CSF or GM-CSF due to low expression of CSF-1 and CSF-2 receptors (150). From those monocytes that could differentiate into macrophages in these patients, they had significant reduction in hemoglobin-haptoglobin receptor CD163 and Tie2 angiopoietin receptor, which are expressed on M2-like macrophages responsible for tissue regeneration (150). In addition, high levels of adipokine leptin detected in blood of hemophiliacs has been shown to inhibit this M2-like regenerative phenotype in human THP-1 monocyte cells (150). Another study using human monocyte-derived macrophages showed that the use of recombinant FVIII Fc fusion protein, not FVIII, uniquely skews macrophages toward an M2-like repair response (151). Together, this suggests that FVIII may favor M2 polarization to suppress proinflammatory response and enhance profibrotic and tissue repair in vitro and ex vivo. However, in vivo studies are needed to validate and further elucidate FVIII role in macrophage polarization.

During tissue injury, platelet-derived thrombin recognized by PAR-1 on macrophages triggers TF production (153). This creates a positive feedback loop where the FIIIa:FVIIa complex binds to macrophages’ PAR-2 resulting in M1-like responses including ROS and proinflammatory cytokine production (154). Interestingly, it has been shown that type 1 interferon (IFN) signaling is required for macrophages’ TF production, which induced caspase-11 to activate inflammasomes for pyroptosis (inflammatory cell death) (155, 156). Inflammasome activation by anti-viral/bacterial type 1 IFN- or LPS-induced signaling leads to the release of microvesicles containing TFs, triggering coagulation (156, 157). These findings suggest an important link that bacterial and viral sepsis can trigger excessive inflammatory M1-like responses leading to disseminated intravascular coagulation (DIC), a life-threatening condition observed in patients with severe COVID-19 and bacterial infections (155, 158–161).

Macrophages express receptors and integrins that recognize thrombin, Fg, and fibrin, these three factors can also modulate macrophage polarization to promote proper wound healing. Although not limited to these receptors, TLR-4, Mac-1 (ɑ_Mβ₂ or CD11b/CD18), and CD11c/CD18 (ɑ_Xβ₂) on macrophages can recognize Fg and fibrin (14, 162–165). In a study using murine and human macrophage cell lines, soluble Fg interaction with TLR-4 stimulated proinflammatory chemokines, specifically macrophage inflammatory protein (MIP)-1ɑ, -1β, -2, and monocyte chemoattractant protein-1 (MCP-1) (166). Interestingly, in an in vivo study that used mice expressing Fg with a mutated γ chain (Fgγ^390-396A) showed that macrophages, monocytes, and neutrophils failed to interact with the mutated Fg, resulting in a major defect in the inflammatory response upon S. aureus peritoneal infection and leading to inability to rapidly clear the infection (14, 167). Although both Fg and fibrin are important for promoting antimicrobial host response, there is a need to elucidate the differential role of soluble Fg versus fibrin on macrophage response. To distinguish the effects of Fg and fibrin on host response, Prasad and colleagues generated Fg^AEK mice that display normal levels of circulating Fg but lack the ability to form fibrin clots due to a germ-line mutation in the Fg’s Aɑ chain thrombin cleavage site (168). They found these mice were unable to clear S. aureus in the peritoneal cavity within the first hour similar to Fg-deficient mice, suggesting the importance of fibrin polymer in inducing antimicrobial response during acute response (168). However, the exact mechanism of how soluble Fg interacts with immune cells, specifically macrophages, in vivo and its antimicrobial response to prolonged infection to other pathogens and tissue injury warrants further studies.

An ex vivo study showed that soluble Fg-induced murine BMDMs to upregulate iNOS and produce proinflammatory cytokines, particularly TNF-ɑ, MCP-1, CXCL9, MIP-1ɑ, MIP-2, and CCL2 (169). Importantly, fibrin outcompeted soluble Fg to promote an M2-like anti-inflammatory phenotype by upregulating Arg1 and secreting IL-10 (169). From this study, it is unclear whether soluble Fg and fibrin favors interactions with TLR-4, Mac-1, or CD11c/CD18 on macrophages. Factors such as tethering properties, protein conformation, and binding site presentation of soluble Fg and fibrin can influence specific interaction with macrophage receptors and integrins, which may subsequently drive distinct pro- or anti-inflammatory intracellular signaling pathways. Thus, it is important to elucidate the mechanisms of how Fg and fibrin can differentially alter macrophage polarization and intracellular signaling pathways.

Fibrinolysis and RAMPs promote M2-like macrophages to restore tissue homeostasis

During the tissue repair and restoration process, M2-like macrophages play an important role in resolving inflammation, repairing, remodeling, and restoring damaged tissue. Besides cytokines, PAMPs, DAMPs, and macrophages can also be polarized by resolution-associated molecular patterns (RAMPs) and coagulation factors secreted from damaged tissue (170). During tissue repair, dying cells release RAMPs, playing a role in transitioning M1s to M2s. Dying cells that release heat shock protein (HSP)-10, HSP27, binding immunoglobulin protein (BiP), and annexin A1 promote autocrine IL-10 production polarizing macrophages toward M2 (171). In addition, phosphatidylserine flipped to the dying cells’ surface activates Mer tyrosine kinase (MerTK) receptors on macrophages and drives M2-like responses such as efferocytosis (172). These RAMPs contribute to their role in promoting anti-inflammatory M2 response [discussed in detail by Koncz et al. (173)].

In Alzheimer’s disease, multiple sclerosis, and experimental autoimmune encephalomyelitis mouse models, multiomic profiling of fibrin-induced microglia revealed that fibrin-CD11b signaling promoted proinflammatory oxidative stress and type 1 IFN response, which led to neuronal dysfunction (174). Furthermore, genetic deletion of Fg binding motif to CD11b in these diseased mice models reduced microglial neurodegenerative signatures, suggesting that targeting excess fibrin can favor clinical outcomes for patients with neurodegenerative or autoimmune diseases (174). As part of fibrin clot dissolution, CCR2⁺ M1-like murine macrophages possess the ability to clear fibrin and fibrin degradation products (FDPs) via endocytosis and lysosomal degradation dependent on plasminogen (PG) and plasmin activity (175). Interestingly, when fibrin gels were subcutaneously implanted in PG-deficient mice, there was a reduced presence of M1 macrophages with uptake of extravascular fibrin (175). Furthermore, they found that fibrin degradation by M1s via this plasmin-dependent endocytic mechanism, even in the absence of all three Fg receptors Mac-1, intercellular adhesion molecule-1 (ICAM-1), and mannose receptor (MR) (175). Based on this study, these M1 macrophages express their cell-surface PG receptor for PG binding, which then gets activated by cell-surface uPA and tPA (175). This leads to plasmin to partially degrade fibrin extracellularly and enable endocytosis of these FDPs by M1 macrophages to fully undergo complete degradation within the lysosomes (175). Other in vitro and in vivo mouse studies also reported that PG is essential in promoting macrophages’ phagocytosis of apoptotic thymocytes, hepatocytes, and alveolar neutrophils, which may be explained by upregulating expression of receptors that recognize RAMPs from dying cells (170, 176–178). Hence, this suggests that PG has an important role in resolving inflammation by initiating M1 macrophages to degrade fibrin both extra- and intracellularly and promote efferocytosis.

Since fibrin serves as a provisional ECM before being replaced by collagen, it may be possible that PG stimulates M1s to initiate the efferocytosis and fibrin degradation before transitioning into different M2 phenotypes to complete ECM remodeling and restoration. RAMPs released from dying cells can polarize macrophages toward different M2 phenotypes (170). In the cell proliferation phase, wound-healing macrophages elevate the production of growth factors PDGF, vascular endothelial growth factor (VEGF), and TGF-β to promote cell proliferation, granulation tissue/ECM formation, and angiogenesis (179). At the same time, neutrophils that phagocytize apoptotic cells induce another subset of M2 macrophages to produce IL-10 for suppressing proinflammatory activities (180). In the remodeling phase, pro-resolving macrophages are activated, allowing the production of matrix metalloproteases (MMPs) to remodel the ECM and restore tissue homeostasis (52). Together, a proper orchestration of macrophage polarization by coagulation factors is critical for tissue homeostasis since dysregulated polarization may result in exacerbated inflammation, resulting in chronic inflammation and promoting the onset of inflammatory diseases including atherosclerosis, diabetes, autoimmune diseases, and coagulopathy-associated diseases (87, 181–186).

PERSPECTIVES AND CONCLUSIONS

Macrophage polarization has emerged as a key feature in gaining a better understanding of how macrophages promote or resolve infections, inflammation, and cancers. One of the challenges to understanding macrophage function is that the description of macrophage polarization is disputed, oversimplified, and confusing due to the lack of nomenclature and experimental standardization. This is an issue that has been persistently brought to attention among macrophage and phagocyte biologists for at least two decades with great debate (32, 65, 187, 188). The use of in vitro and ex vivo studies with macrophage-like cell lines, murine bone marrow-derived or peritoneal macrophages, and human macrophages isolated from peripheral blood allowed researchers to stimulate these macrophages with various inducers generating different phenotypes (32). Since each laboratory varies in defining macrophage activation and polarization states, it is important to provide reporting standards for these studies such as (i) describing the source and differentiation method of macrophages; (ii) defining the activator with the recommended M(inducer) notation (i.e., M(IFN-γ), M(LPS), M(IL-4), M(GCs), etc.); (iii) reporting activation conditions; and (iv) stating the method of analysis and readout (32). While in vitro models demonstrate induction of discrete macrophage polarization states (M1 vs M2), in vivo disease models have revealed that polarization occurs on a continuum with mixed phenotypes due to the complex TME. Therefore, it is difficult to translate in vitro and ex vivo characterization to in vivo mouse models and human diseases due to other underlying factors that can also polarize diverse macrophage populations in these scenarios. Hence, there is a great need to go beyond just detection for the presence or absence of markers to characterize macrophages in complex TMEs. A proposed solution is to elucidate the interplay of epigenetics and metabolism in regulating and characterizing macrophage polarization. Recent studies and literature reviews shed a light on the epigenetics and metabolism on driving macrophage polarization and function marking advancements in this direction (189–193). The challenge to studying this is that it requires the use of multiple specialized techniques and instruments with its own set of limitations and difficult technical expertise. One group addressed this obstacle by developing a novel triomics (epigenetics, metabolomics, and transcriptomics) method using mass spectrometry on IFN-γ- or IL-4-stimulated BMDMs to simultaneously analyze metabolites, histone modifications, and protein expression all in a single sample (194). This novel method is one of many tools that can be applied to in vivo and clinical samples, thereby generating detailed and meaningful characterization and functions of macrophages in different diseases. While there is much progress in understanding macrophage polarization, there is still the need to further develop approaches to better understand how the host tissue microenvironment in inflammation, infections, and cancers affect macrophage functions.

Specifically, an area that needs to be investigated further is how the activation of the coagulation cascade upon tissue injury influences macrophage functions during infections, since it has been reported that a major component of this cascade, Fg and its polymerized form fibrin, can modulate macrophage polarization and antimicrobial responses to infections and insertion of medical devices (65, 195, 196). This is relevant in clinical settings, specifically when using implanted biomaterials, such as urinary catheters. These implants induce tissue damage, which activates the coagulation cascade, resulting in Fg recruitment and its polymerization to fibrin clot by thrombin proteolytic activity (197, 198). In catheter-associated urinary tract infections (CAUTIs), urinary catheterization elicits bladder inflammation in both mice and humans, promoting Fg/fibrin accumulation and recruitment of neutrophils and macrophages into the catheterized bladder (199–205). Importantly, in humans and mice, Fg and fibrin are used as platforms for biofilm formation on the catheters and bladders by diverse uropathogens including E. coli, E. faecalis, C. albicans, K. pneumonia, P. aeruginosa, S. aureus, and Acinetobacter baumannii (199–205). Furthermore, CAUTI studies performed on mice with fibrinolytic deficiencies revealed that mutated soluble Fg decreased microbial burden while excessive fibrin accumulation enhanced microbial colonization (198). The fact that catheterization allows different microbes to infect is puzzling given the strong immune cell recruitment, indicating that the immune response generated is unable to control the infection. Since Fg and fibrin are accumulated in the catheterized bladder, they could play a role in macrophage polarization. It is known that macrophages integrins (Mac-1 [ɑ_Mβ₂] and ɑ_Xβ₂) and TLR-4 interact with Fg, resulting in M1 polarization and proinflammatory responses (14, 162, 167). Oppositely to Fg, fibrin promotes M2 polarization and prevents M1 polarization by Fg, LPS, and IFN-γ (169). This differential polarization could be explained by differences in the tethering properties, receptor or integrin engagement, and protein conformation between Fg and fibrin. Coagulation factors that promote fibrin formation could also indirectly induce M2 polarization in other diseases such as cancer (206, 207). For example, it was reported that higher activity of FX, a plasma protein that activates prothrombin into thrombin for fibrin formation, dampened antitumor response by inducing M2 polarization in cancer mouse models (206, 207). Interestingly, a study found that chemical inhibition of FX using rivaroxaban in myeloid cells reprograms M2 tumor-associated macrophages to M1 that promote antitumor immunity (207). Therefore, it is important to further study the role of Fg and fibrin on macrophage polarization in different disease and infection models to better understand and develop reprogramming strategies.

A major challenge is to predict how the complex inflammatory microenvironments will affect macrophage polarization and their clinical associations with disease pathologies. Since macrophage polarization occurs on a spectrum, there is a major need to elucidate mixed M1/M2 phenotypes. Although great strides have been made in developing novel strategies targeting macrophage polarization, still these therapeutics have shown mixed results in different disease models. This could be explained by how other factors in the TME can affect both macrophage and other immune cell responses. Therefore, understanding how the TME drives the plasticity of macrophage polarization by modulating pro- and anti-inflammatory phenotypes can provide an avenue to developing viable therapeutic strategies in human diseases.

Biographies

Armando M. Marrufo graduated with his Ph.D. in the Biological Sciences at the University of Notre Dame, where under mentorship of Dr. Ana Flores-Mireles his research elucidated how the urinary catheterized bladder environment induced anti-inflammatory macrophage polarization to promote catheter-associated urinary tract infections. Prior to his graduate studies, he completed his Bachelor of Science degree in Biological Sciences at Texas A&M University in College Station, Texas, where his first research experience focused on elucidating the mechanism of how bacteriophages induce lysis of gram-negative bacteria. He then went on to complete his Master of Science degree in Biomedical Sciences at the University of North Texas Health Science Center in Fort Worth, Texas, where his research aimed to utilize natural killer cells to target inhibitory ligands on triple-negative breast cancer cells.

Anna Lidia Flores-Mireles did her B.S. in Marine Biology and M.Sc. in Marine Biotechnology in Mexico. She did a PhD in Microbiology at Cornell University under Dr. Steve Winans' supervision. Then, she did her postdoctoral research at Washington University School of Medicine under the supervision of Dr. Scott Hultgren and Dr. Michael Caparon. In 2018, she became the Janet C. and Jeffrey A. Hawk Assistant Professor of Biological Sciences at the University of Notre Dame. Her laboratory is working on understanding how catheter-induced bladder inflammation renders the host susceptible to catheter-associated urinary tract infections and subsequent systemic dissemination. By utilizing a comprehensive approach, combining microbial genetics, transgenic mice, clinical samples, material modification, pathogen and host transcriptomics and proteomics as well as immunological, histological, microscopic, and pharmacological techniques, her lab goals are to develop novel antibiotic-sparing therapies against these infections.