Macrophage polarization in chronic kidney disease: a balancing act between renal recovery and decline?

Abstract

Macrophages are heterogenous cells of the innate immune system that can fluidly modulate their phenotype to respond to their local microenvironment. They are found throughout the renal compartments, where they contribute to homeostasis and function. However, renal injury activates molecular pathways that initially stimulate differentiation of macrophages into a proinflammatory M1 phenotype. Later in the course of healing, abundant apoptotic debris and anti-inflammatory cytokines induce the production of anti-inflammatory M2 macrophages, which contribute to tissue regeneration and repair. Thus, the dynamic balance of M1 and M2 populations may outline the burden of inflammation and process of tissue repair that define renal outcomes, which has been the impetus for therapeutic efforts targeting macrophages. This review will discuss the role of these phenotypes in the progression of chronic renal injury, potential pathogenic mechanisms, and the promise of macrophage-based therapeutic applications for chronic kidney disease.

INTRODUCTION

Inflammation is a prominent feature of chronic kidney disease (CKD) and plays a leading role in the progression of renal injury and increased cardiovascular mortality of these patients (42). Various effectors, including renal parenchymal release of damage-associated molecular patterns (40) (DAMPs), upregulation of proinflammatory transcription factors such as NF-κB (1) and subsequent proinflammatory cytokine pathways (44) [e.g., TNF-α and chemokine (C-C motif) ligand 2], and renal infiltration by immune cells (26), are responsible for the increased inflammation in CKD.

Macrophages are members of the innate immune system that arise from circulating monocytes or from populations of self-renewing resident tissue macrophages (17). Although they were initially characterized by their phagocytic activity (15), an equally important function of macrophages is their ability to regulate tissue homeostasis and promote immune tolerance (16). Macrophages change their phenotype in response to their local environment, leading to the development of a classification scheme used in vivo and in vitro, as described elsewhere (43). Briefly, macrophages are described as M1 or M2. Although this classification scheme may be overly simplistic (35), given that macrophage polarization is akin to a spectrum with M1 and M2 on opposite ends, the M1-M2 phenotype balance can provide valuable insights into the inflammatory status and pathophysiology of renal disease.

In the normal response to injury, damaged tissue releases DAMPs, leading to the differentiation of macrophages to the M1 phenotype, which is characterized by the release of proinflammatory cytokines and increased microbicidal and tumoricidal activity (24, 32, 39). Days to weeks after the initial injury, macrophages repolarize to the M2 phenotype, defined by release of anti-inflammatory mediators, growth factors, and proangiogenic cytokines to facilitate the healing process (24, 32, 39). Macrophages have gained attention in recent years for their role in the progression to CKD. In the kidney, M1 macrophages are considered deleterious, as they sustain the proinflammatory environment, leading to the progression of renal injury and development of fibrosis (19, 21, 26). In contrast, M2 macrophages are associated with renal repair and functional recovery (20, 32). This review will highlight mechanisms by which these cells may contribute to the pathogenesis of chronic renal diseases. Furthermore, we will briefly discuss the exciting potential of targeting macrophages as a therapeutic strategy for kidney disease.

MACROPHAGE POLARIZATION AND THE KIDNEY

Mounting evidence shows the capacity of renal macrophages to fluidly switch to either the proinflammatory or tissue-healing phenotype, depending on the microenvironment (27, 31, 39). The mechanisms underlying the switch from the M1 to M2 phenotype are multifactorial and not fully understood and are briefly discussed below.

Inflammatory Macrophages and Renal Injury

M1 macrophages act as first responders against pathogens and activate the immune system (43). However, their broad-reaching, rapid response comes at the cost of target specificity. Indeed, factors produced by M1 macrophages, such as TNF-α (46) and reactive oxygen species (3), may cause endothelial dysfunction (30) and death of healthy renal cells (51). If unchecked, persistent M1 infiltration and associated inflammation contribute to the decline of renal function (2, 11) and, ultimately, fibrosis (27, 36).

In the normal course of renal injury and repair, initiation of the healing response attenuates the release of DAMPs from the renal parenchyma. It is hypothesized that this downward trend in DAMP production may play a role in the shift to the M2 phenotype, as DAMPs are strong signals for maintaining M1 polarization (38). However, some renal recovery is required to initiate this response, as progressive injury leads to the sustained release of DAMPs and M1 polarization [as shown in CKD (6) and ischemia-reperfusion injury models (26)]. On the other hand, acute depletion of macrophages after renal injury, during the M1-predominated response, has been shown to be beneficial in immunologic-, toxin-, ischemia-, and hypertension-mediated renal injury models (8, 14, 21, 47). A recent study on novel mouse knockout of IL-1 receptor-associated kinase M, a macrophage-specific inhibitor of Toll-like receptor and IL-1 signaling, showed that this single knockout significantly increased M1 macrophages in the ischemia-reperfusion injured kidney and contributed to tubular loss, glomerulosclerosis, and interstitial fibrosis (26). Furthermore, IL-1 receptor-associated kinase M knockout prevented renal accumulation of M2 macrophages after acute injury, suggesting a role in the switch from the M1 to M2 phenotype (26). Together, these data suggest that M1 macrophages are active players in the pathogenesis of renal injury and may produce a vicious cycle, where damage leads to the release of DAMPs, which subsequently stimulate Toll-like receptors to further stimulate M1 polarization. Sufficient initial tissue healing to allow breakthrough of the M2 phenotype may thus be an important factor in recovery from acute injury versus progression to CKD.

Trophic Macrophages and Renal Repair

Macrophages have been traditionally seen as injurious mediators in the pathogenesis of kidney disease (22, 41). Indeed, previous studies have shown a protective effect of macrophage depletion acutely after renal injury in multiple models (9, 14, 21, 23, 45). However, other studies have shown that depletion of macrophages in the subacute stages of renal injury, during the peak of M2 infiltration, may exacerbate tubular damage and fibrosis, supporting the regenerative role of M2 macrophages (12, 27, 37). These disparate effects suggest that the timing for macrophage removal may be a crucial component of such strategies to boost renal repair.

Recent studies uncovering underlying mechanisms of M2-induced renal repair have identified a few promising candidates. One example is Wnt7b, a ligand of the Wnt pathway that drives the formation of the tubular epithelium during development. Wnt7b is reexpressed by macrophages after renal injury and signals through tubular cells to initiate cell proliferation and repair of the tubular basement membrane (28).

Fibrosis is a well-established hallmark of CKD, one that is hypothesized to be indicative of irreversible loss of function, as fibrotic tissue may physically obstruct microvascular proliferation and tubular regeneration (10). Mannose receptor C type 2 (MRC2), a fibrolytic cell surface receptor expressed on M2 macrophages, binds and internalizes collagen. MRC2-deficient mice developed worse fibrosis in the unilateral ureteral obstruction model (29). A recent study (26) has also supported the conclusion that M1 macrophages directly contribute to development of fibrosis by sustaining the renal inflammatory milieu and showed that when inflammation is suppressed, M2 macrophage infiltration does not associate with fibrosis, regardless of disease duration. Together, these findings suggest that the switch to M2 macrophages involves alteration of multiple pathways, leading to a trophic, anti-inflammatory effect on the renal parenchyma. Nevertheless, the net effect of M2 macrophages on fibrosis is still an area in need of research, as some studies have implicated chronic M2 infiltration in facilitating renal fibrogenesis where inflammation is severe (4, 22, 41).

T Cells and Macrophage Polarization

T cells of the adaptive immune system act as both controllers and effectors of macrophage function in CKD. T cells have been shown to increase macrophage recruitment into the kidney, which is a driving mechanism of injury in immune-mediated forms of kidney disease (18). T cells do not appear to directly influence M1/M2 polarization, as this differentiation has been shown to occur in recombination-activating gene-deficient mice (33). Conversely, M2 macrophages have been shown to induce regulatory T cell differentiation both in vitro and in vivo, resulting in reduced injury and fibrosis in an adriamycin-induced nephropathy model (5). The interplay between the adaptive immune system and macrophage polarization is a complex topic that warrants further study.

MACROPHAGE-BASED THERAPEUTICS FOR KIDNEY DISEASE: AN OPPORTUNITY

Myriad studies have shown both pathogenic and protective roles of macrophages in kidney disease of diverse etiology (7, 25, 27, 41), suggesting that macrophages may serve as targets to ameliorate renal injury via manipulation of their polarization.

Genetic Manipulation

Genetically engineered macrophages have been developed to modulate their effector function. Macrophages transduced with the gene for an IL-1 receptor antagonist reduced glomerular and tubular fibrosis in unilateral ureteral obstruction and anti-glomerular basement membrane glomerulonephritis models (52). Another approach is use of adenoviral transduction to induce expression of IL-4 or IL-10 (anti-inflammatory cytokines associated with M2 polarization) in bone marrow-derived macrophages (50). The macrophages were then transferred into kidneys of rats with nephrotoxic nephritis, where they homed to inflamed glomeruli, suppressed immune cell infiltration, and reduced the degree of glomerular injury (50). Finally, genetic manipulation of macrophages to induce expression of IκB, which inhibits nuclear translocation of NF-κB, reduced inflammatory IL-12, TNF-α, and nitric oxide signaling in classically activated M1 macrophages. This was associated with attenuated fibrosis, less crescent formation, and improved albuminuria in a nephrotoxic nephritis model (49).

A major advantage of genetic approaches is the induction of long-lasting changes in phenotype and cytokine expression in environments that would normally favor inflammatory polarization. Furthermore, natural chemotactic triggers cause macrophages to travel to areas of injured renal parenchyma, allowing the induced gene product to be secreted directly at the site of damage. However, major hurdles must be overcome before this therapeutic strategy is translated to clinical use; chief among these are ensuring the safety of viral transduction and reliably controlling the degree of gene suppression or induction.

Cytokine Manipulation

Administration of cytokines is a relatively straightforward method of modulating macrophage phenotype. Macrophages isolated from mouse spleens and exposed to IL-4 and/or IL-13 in vitro to induce the M2 phenotype were injected in an adriamycin-induced nephropathy model (48), resulting in reduced expression of proinflammatory cytokines and attenuated inflammatory markers in renal immune cells. Similar improvements have been observed after administration of M2 macrophages in models of ischemia-reperfusion injury and diabetic-induced renal injury (27). In tumor models, VEGF signaling through VEGF receptor type 1 has been shown to promote macrophage infiltration and differentiation into an angiogenic M2-like phenotype (34). Furthermore, our laboratory has recently found in a swine model of CKD (6) that renal therapy using a biopolymer-stabilized form of VEGF significantly shifts the phenotype of renal-infiltrating macrophages from indolamine 2,3-dioxygenase-expressing M1 to MRC1-expressing M2 without reducing the total number of macrophages (13). Interestingly, these M2 macrophages densely express VEGF and are associated with improved renal hemodynamics, microvascular integrity, and attenuation of renal fibrosis, suggesting that this phenotypic change after VEGF therapy may play a role in renal recovery in CKD (13). Future studies of macrophage depletion and reconstitution may be required to further establish whether a renal shift to the M2 phenotype is feasible, translatable, and functionally significant as a therapeutic strategy.

CONCLUSIONS

Regardless of etiology, renal parenchymal injury induces the release of chemotactic factors, renal infiltration of macrophages, and polarization to the M1 phenotype. M1 macrophages are characterized by the production of inflammatory cytokines and, if unopposed, may lead to tubular cell death, endothelial dysfunction, inflammation, and fibrosis while suppressing M2 repolarization. Thus, a sustained abundant M1 infiltration can contribute to the progression of renal injury (Fig. 1). Promising strategies to alter effector phenotype at the site of renal injury via genetic or cytokine manipulation of macrophages are being developed. However, current evidence warrants additional studies to better understand the specific roles and niches of macrophage phenotypes and their signaling pathways to unravel their translational application as therapeutic strategies to induce renal recovery.

Fig. 1.

A schematic overview of macrophages in progression of renal injury. The initial renal damage leads to the release of chemotactic factors, inducing macrophage infiltration and subsequent differentiation to the M1 phenotype. M1 macrophages release multiple inflammatory cytokines, which drive tubular injury, endothelial dysfunction, and inflammation, leading to fibrosis and progression of renal injury toward chronic kidney disease (CKD) if left unchecked. In the normal course of healing, macrophages shift to the M2 phenotype and participate in the repair process through production of VEGF and Wnt7b (favoring microvascular recovery and parenchymal regeneration respectively, red lines) and surface expression of mannose receptor C type 2 (MRC2) to reduce the progression of fibrosis (red lines) and stimulate fibrolysis. M2 differentiation is favored by polarizing factors produced by M1 macrophages, such as IL-1 receptor-associated kinase M (IRAK-M), and inhibited by persistent inflammation favoring M1 polarization. Modulation of macrophage function or polarization state by genetic or cytokine approaches (*potential therapeutic targets) offers new potential therapeutic strategies for improving outcomes in kidney disease. DAMP, damage-associated molecular patterns.

GRANTS

This work was supported by National Institutes of Health Grants R01-HL-095638, P01-HL-51971, and P20-GM-104357 and American Heart Association Grants IPA3417016 and PRE34380274.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.E.E. and A.R.C. conceived and designed research; J.E.E. prepared figures; J.E.E. drafted manuscript; A.R.C. edited and revised manuscript; J.E.E. and A.R.C. approved final version of manuscript.