OUP user menu

★ Editor's choice ★

Cardioprotective function of cardiac macrophages

Katsuhito Fujiu , Jack Wang , Ryozo Nagai
DOI: http://dx.doi.org/10.1093/cvr/cvu059 232-239 First published online: 27 March 2014


The heart is composed of several cell types including cardiomyocytes, cardiac fibroblasts, endothelial, and smooth muscle cells. In addition to these major cell types, cardiac macrophages are also present in small numbers under physiological conditions. Recently, the resident macrophage is considered to have vital functions in the maintenance of tissues and homeostasis in many organs, including brain, liver, adipose tissue, lymphatic tissue, and intestinal tract. However, detailed functions of the cardiac resident macrophage are not fully understood. Although the removal of debris arising from damaged cardiomyocytes and pro-inflammatory effects after heart injuries are conventional tasks of cardiac macrophages (classically activated macrophage or M1 macrophage), novel functions like anti-inflammatory roles, adaptive response, and tissue maintenance have also been reported in recent years. Macrophages that possess these novel functions are generally so-called M2 macrophages, which are alternatively activated and show anti-inflammatory phenotype under pathological conditions. In this review, we focus on the cardioprotective function of the cardiac macrophage and discuss in light of unveiled fundamental functions of macrophages that have been also found in other organs.

  • Macrophage
  • Heart
  • Heart failure
  • M2
  • Alternative activated

1. Introduction

The heart is composed of several cell types including cardiomyocytes, cardiac fibroblasts, endothelial cells, smooth muscle cells, and immune cells including cardiac macrophages.1 Recent studies have shown that macrophages are composed of heterogeneous subpopulations with various functional properties.2,3 While haematopoietic stem cells were thought to be the exclusive origin of macrophages, recent studies have reported the existence of yolk sac-derived macrophages,46 which improve our understanding and accelerate further investigations on macrophages. In addition to removing debris from the tissue, macrophages also have various other roles within the tissue such as contributing to the maintenance of tissue and organ homeostasis via various mechanisms.7,8 In the current review, we focus on the function of macrophages in the heart under pathophysiological conditions and discuss the cardioprotective functions of cardiac macrophages.

2. Protective role of the cardiac macrophage under pathological conditions

2.1 Effects of macrophage depletion by clodronate liposome administration

In postnatal life, when the myocardium requires tissue remodelling with high numbers of macrophages such as during myocardial infarction, Ly-6Chi monocytes are recruited from the bone marrow and spleen and appear to become monocyte-derived macrophages.9,10 On the other hand, functional contribution of organ resident macrophages under pathological conditions has been reported in atherosclerosis, in which bone marrow- and spleen-derived macrophages are major contributors only at the early disease stage, and resident (lesional) macrophages become dominant at later phases of arteriosclerosis formation.11 Resident macrophages are expected to have functional properties distinct from that of monocyte-derived macrophages.3,12 However, there are no previous studies that use highly sophisticated methods—such as specific resident macrophage depletion or specific resident macrophage gene deletion—to characterize the functional properties of resident macrophages in the heart. In general, it is technically difficult to perform specific gene manipulations of resident cardiac macrophage. Therefore, one of the currently available methods is to use clodronate, which binds intracellular ATP and inhibits ATP function, resulting in cellular apoptosis.13 Encapsulation of clodronate with liposome enables us to selectively deliver the clodronate into macrophages via their phagocytic activity and thereby depletes the cardiac macrophages. Repeated treatments with clodronate liposome can also deplete inflammatory monocytes/macrophages from the bone marrow and in peripheral blood.11,14 However, the method cannot distinguish functional differences between resident macrophages and monocyte-derived macrophages. To observe the function of resident macrophages separate to monocyte-derived macrophages, we can apply a single injection of clodronate liposome before the creation of disease models, or use disease models in which resident macrophages can dominantly contribute.

Several studies have reported cardioprotective effects of cardiac macrophages using the clodronate liposome in rodent models. Administration of the clodronate liposome into hypertensive rats (Ren2 rat) reduced the number of cardiac macrophages and induced CD4+ T-cell-dominant inflammatory cell infiltration15 (Figure 1). The cardiac contractility was decreased in the early period by macrophage depletion. These results suggest that cardiac macrophages can protect myocardium against hypertension-induced stress responses15 (Figure 1). In a mouse myocardial cryoinjury model, the clodronate liposome administration delayed removal of debris from myocardium and decreased neovascularization, resulting in ventricular dilatation and a high mortality rate16 (Figure 1). Several studies showed that vascular endothelial growth factor alpha (VEGF-A) and transforming growth factor beta (TGF)-β mediated debris removal by macrophages and neovascularization.1719 Consistent with these reports, VEGF-A and TGF-β expressions were decreased in the clodronate liposome-treated cryoinjured myocardium, suggesting that cardiac macrophages preserve cardiac contraction and healing after myocardial infarction at least in part through releasing VEGF-A and TGF-β16 (Figure 1). Given these results, cardiac macrophages are considered to supply protective effects against cardiac stress when methods involving the global depletion of cardiac macrophages are applied.

Figure 1

The cardioprotective role of cardiac macrophages. Cardiac macrophages have been classified into two groups under pathophysiological conditions, Ly-6ChiCD206CD204 classically activated (M1) macrophages and Ly-6CloCD206+CD204+ alternatively activated (M2) macrophages. Clodronate liposome-based macrophage depletion affected all kinds of macrophages. M1 macrophages have a functional role under conditions of myocardial damage such as myocardial necrosis and/or debris formation post-myocardial infarction, during which debris clearance is mediated through TGF-β  and neovascularization through VEGF-A. The M1 macrophage activates CD4 T cells via IL-12 secretion, and cell–cell interactions between activated CD4 T cells and M1 macrophages induce a pro-inflammatory phenotype in the macrophage. On the other hand, M2 macrophages inhibit CD4 T cell and granulocyte activity and exhibit anti-inflammatory properties and activity against excessive fibrosis. M2 macrophage development is regulated by SGK1 and/or IL-13, but not IL-4. STAT3 is required for differentiation of M2 macrophages. M2 macrophages secrete IL-12, which activates CD4 T cells. Cell–cell interactions between M2 macrophages and activated CD4 T cells are also required for M2 macrophage differentiation and proliferation during cardiac stresses. M2 macrophages are functional in cardiac hypertrophy and the myofibroblast-mediated healing response to pressure stress or hypertensive stress. Appropriate cardiac remodelling mediated by M1 and M2 macrophages results in the preservation of cardiac function and survival. Imbalances in the function of these two macrophage types induce unfavourable cardiac remodelling. VEGF-A: vascular endothelial growth factor alpha; TGF-β: transforming growth factor beta; SGK1: serum- and glucocorticoid-inducible kinase 1; IL: interleukin; STAT3: signal transducer and activator of transcription 3.

2.2 Functional diversity of cardiac macrophages by M1/M2 polarization balance

Although macrophage heterogeneity might be complex and not still fully understood, functional binary classifications of macrophages have been used for easy understanding of this field, such as the M1/M2 categorization or classically activated/alternatively activated macrophage in sterile inflammation. Several reports have addressed the mechanisms by which these different kinds of macrophages handle tissue remodelling after tissue injury by using genetic manipulation of polarity-determining genes, as described below.

2.2.1 Alternative macrophage activation by Th2 cytokines

Since interleukin (IL)-4 or IL-13 induces M2 phenotype macrophage differentiation in bone marrow-derived macrophages in vitro,2023 IL-4 and IL-13 may be also related to the alternative activation (M2 activation) of resident macrophages and newly recruited monocyte-derived macrophages in vivo. Several studies of viral infection or immune peptide-induced myocarditis models have investigated for IL-4 and IL-13 involvement in M2 activation. The Il13 knockout mouse developed severe myocarditis and heart failure in which numbers of CD206- and CD204-positive alternatively activated macrophages decreased and CD4+ T cells, CD45 leucocytes (mainly granulocytes), and fibroblasts were increased within the myocardium24 (Figure 1). Increase in inflammatory cytokines, such as IL-1β, IL-18, and interferon-γ, and other factors promoting myocardial fibrosis, such as TGF-β, may also have contributed to cardiac functional deterioration and myocardial fibrosis (increase in CD4 response)24 (Figure 1). In contrast, the Il4 knockout mouse did not show any significant differences in the development of myocarditis compared with wild-type littermates.24 These findings suggest that IL-13, rather than IL-4, dominantly regulates macrophage polarity in myocarditis and the alternative activation of macrophages, thereby resulting in anti-inflammatory and cardioprotective effects24 (Figure 1).

2.2.2 Serum- and glucocorticoid-inducible kinase 1

Serum- and glucocorticoid-inducible kinase 1 (SGK1) is a key mediator of fibrosis. Cardiac hypertrophy and fibrosis induced by angiotensin II infusion did not develop in Sgk1 knockout mice, but developed in wild-type mice, despite equivalent hypertensive levels in both groups.25 Induction of cardiac M2 macrophages and leucocytes by angiotensin II were diminished by SGK1 deficiency25 (Figure 1). Phosphorylation and nuclear translocation of signal transducer and activator of transcription 3 (STAT3), a key regulator of alternative macrophage activation, were activated by angiotensin II administration in wild-type mice.25 However, SGK1 deficiency attenuated STAT3 activation and alternative activation of macrophages in an in vitro model,25 suggesting that SGK1 induced cardiac fibrosis at least in part through M2 macrophage proliferation and activation (Figure 1).

2.2.3 Interleukin-12

In contrast, the production of IL-12 (IL-12p35), which is essential for the activation of interferon-γ-producing T cells, by cardiac macrophages was also induced after angiotensin II infusion.26 Increased excessive induction of M2 macrophages due to deficiency of IL-12 was observed, which resulted in markedly enhanced cardiac fibrosis in the same angiotensin II infusion mouse model compared with wild-type mice.26 Although deficiency of IL-12p35 did not affect macrophage differentiation in a cell-autonomous manner, IL-12 produced by cardiac macrophages affected CD4+ T cells. Moreover, cell–cell interactions between IL-12-activated CD4+ T cells and cardiac macrophages shifted the macrophage polarity towards the M1 phenotype and inhibited excessive M2 macrophage proliferation (Figure 1). This report also indicated that the IL-12-mediated shift of polarity towards the M1 phenotype prevents excessive cardiac fibrosis.26

2.2.4 Macrophage scavenger receptor class A

Macrophage scavenger receptor class A (SR-A), a key modulator of inflammation, has also been reported to shift the cardiac macrophage polarity towards the M2 phenotype27 (Figure 2). In a mouse myocardial infarction model, SR-A-knockout mice showed exacerbated cardiac systolic function and fibrosis in the myocardial infarction model compared with wild-type mice. M2 to M1 macrophage polarity shift was observed in SR-A-knockout mice after myocardial infarction, and pro-inflammatory cytokine levels, including IL-1β, IL-6, and tumour necrosis factor alpha (TNF-α), were augmented27 (Figure 2). These phenotypes in SR-A-knockout mice were partially rescued by bone marrow transplantation from wild-type mice. SR-A on cardiac macrophages contribute to M2 phenotype, anti-inflammatory, and anti-fibrotic remodelling after myocardial infarction.

Figure 2

Two distinct monocyte-derived cardiac macrophages differently contribute to the healing process against cardiac stress or injury. In the case of acute heart injury such as myocardial infarction, monocytes recruited from bone marrow and spleen are another sources of cardiac macrophages. Ly-6Chi monocytes are recruited to injured myocardium via CCR2 and differentiate into classically activated M1 macrophages. These macrophages produce IL-1β, IL-6, and TNF-α and exhibit phagocytic, proteolytic, and pro-inflammatory functions. M1 macrophages phagocytose and clear debris within the heart in early phases of cardiac injury. In later phases, Ly-6Clo monocytes are recruited to the myocardium via CX3CR1 and differentiate into M2 macrophages, which express TGF-β and IL-12. SR-As on M2 macrophages are essential for the maintenance of the M2 phenotype. M2 macrophages contribute to anti-inflammatory response, neovascularization, and myofibroblast activation during the healing process after cardiac injury. CCR2: CC chemokine receptor 2; CX3CR1: CX3C chemokine receptor 1; SR-A: macrophage scavenger receptor class A.

2.2.5 Mineralocorticoid receptor on cardiac macrophages regulates macrophage polarity

The anti-aldosterone drug spironolactone suppresses cardiac fibrosis28 and reduces cardiovascular events.29,30 Eplerenone, which selectively inhibits the aldosterone receptor, for example, mineralocorticoid receptor (MR), also suppresses cardiac fibrosis31 and reduces cardiovascular events.32,33 A study of the nitric oxide synthase inhibitor NG-nitro-l-arginine methyl ester (l-NAME) and angiotensin II treatment (hypertension model) in a macrophage-specific MR-knockout mouse showed that despite increases in blood pressure and heart rate, cardiac remodelling (cardiac hypertrophy and cardiac fibrosis) and vascular remodelling were suppressed.34 The evidence demonstrates a significant reduction in cardiac M1 macrophages in MR-knockout mice, suggesting that the presence of MR mediates M1 macrophage recruitment (Figure 2). The l-NAME treatment without angiotensin II in the MR-knockout mouse revealed that while cardiac macrophages increased in both wild-type and MR-knockout mice, the ratio of M1-/M2-activated macrophages in the MR-knockout mouse was reduced. The results suggest that a M2 macrophage polarity shift occurred in the knockout mouse and suppressed cardiac fibrosis.27,34 Furthermore, aldosterone treatment of wild-type peritoneal macrophages increased the M1 classically activated pro-inflammatory cytokines, TNF-α, monocyte chemotactic protein-1, regulated on activation, normal T cell expressed and secreted, and IL-12, and accelerated M1 macrophage activation. In contrast, peritoneal macrophages from the macrophage-specific MR-knockout mouse expressed M2 markers at higher levels, and aldosterone treatment did not accelerate M1 activation and inflammatory cytokine expression, suggesting a M2 polarity shift34 (Figure 2). Aldosterone is involved in polarity shift and activation of cardiac macrophages through the MR, thereby contributing to a M1 macrophage polarity shift. Therefore, excessive aldosterone and MR activation may lead to a M1 macrophage-dominant state and suppress M2 polarity shift in response to cardiac stress, resulting in disruption of cardiac macrophage-mediated cardioprotective mechanisms.

Aldosterone binds to both the MR and glucocorticoid receptor (GR), and the DNA-binding motif is shared in both receptors. In the heart, GR expression is significantly higher than MR.35 It is not well understood which of the MR or GR mediates the aldosterone effect. While the MR antagonist, eplerenone, inhibits aldosterone-induced M1 activation of peritoneal macrophages, the GR antagonist fails to inhibit this aldosterone-mediated M1 activation.34 Thus, aldosterone may regulate macrophages via MR-mediated signalling pathways.

2.3 Monocyte-derived macrophages and classically activated macrophages

Bone marrow- and spleen-derived Ly-6Chi monocytes are recruited to the heart during post-infarcted myocardial remodelling via chemokine receptor 2 (CCR2) and become monocyte-derived macrophages.10,14 While monocyte-derived macrophages evoke tissue inflammation continuously, they also have proteolytic and phagocytic effects against debris and apoptotic cells.36 In this model, efferocytosis, which removes debris in the infarcted myocardium, appeared to be suppressed resulting in failure of reverse myocardial remodelling.37 Although inflammatory phenotypic Ly-6Chi macrophages secreted inflammatory cytokines within the infarcted myocardium, it basically acted as cardioprotective macrophages. In contrast, Ly-6Clo monocytes were recruited in the late phase of post-infarcted myocardial remodelling via CX3C chemokine receptor 1, and promoted tissue healing by anti-inflammatory, pro-angiogenesis, and myofibroblast activation effects14,38,39 (Figure 2).

These reports suggest that cardioprotective effects in post-infarcted myocardium are mediated by Ly-6Chi macrophages in the early phase and by Ly-6Clo macrophages in the late phase. Similar investigations were reported for the kidney, where the presence of two types of macrophages, cell apoptosis-induced cytotoxic Ly-6ChiF4/80lo M1 macrophages and TGF-β secreting Ly-6CloF4/80lo alternatively activated macrophages (M2), were reported. Bone marrow-derived Ly-6Chi monocytes were recruited to the kidney and differentiated into M1 macrophages in the early phase of inflammation. In late phases of inflammation, when inflammation was receding, the recruited monocytes differentiated into M2 macrophages. Thus, Ly-6Chi monocytes can differentiate into both M1 and M2 macrophages.40,41 Furthermore, Ly-6Clo monocytes recruited during inflammation only differentiated into M2 macrophages.41 It is therefore likely a universal response that Ly-6Chi macrophages (M1) possess phagocytic and proteolytic activities during the acute inflammatory phase, while Ly-6Clo macrophages (M2) have tissue healing and protective activities during the late inflammatory phase with TGF-β as a key factor.

3. Refined classification of cardiac macrophage subsets

More recently, comprehensive and detailed analyses of mouse cardiac macrophage subsets at the steady state and during inflammation have been reported.42 In that study, CD45+CD11b+F4/80+ macrophages were divided into four subsets by using surface markers, including Ly-6C, major histocompatibility complex class II (MHCII), CD11c, and CCR2 (Figure 3). The Ly-6Chi macrophage was almost consistent with the classically activated M1 macrophage and contributes to inflammation in response to angiotensin II infusion or in myocardial infarction models. On the other hand, Ly-6Clo macrophages, which have been classified as M2 macrophages, were divided into MHCIIhiCD11cloCCR2, MHCIIhiCD11chiCCR2hi, and MHCIIlo macrophage subsets. The MHCIIlo macrophage population sees the largest numerical expansion in cell number after cardiac stress is applied (angiotensin II infusion or myocardial infarction), and these cells have a strong phagocytic activity. The MHCIIlo macrophage plays a role in clearance of dying cardiac cells. Meanwhile, the MHCIIhiCD11cloCCR2 macrophage has an outstanding antigen-processing and -presenting function. However, the number of MHCIIhiCD11cloCCR2 macrophages decreased with cardiac stress. On the other hand, the number of MHCIIhiCD11chiCCR2hi macrophages increased with cardiac stress and also induced the expression of pro-inflammatory cytokines, such as IL-1b via NLRP3 inflammasome. It is particularly worth noting that MHCIIhiCD11chiCCR2hi macrophages, which have been classified as M2 macrophages, show a pro-inflammatory phenotype. Interestingly, pro-inflammatory macrophages (which is almost half of all Ly-6C+ macrophages and MHCIIhiCD11chiCCR2hi macrophages) specifically express CCR2. These results might explain the reason why therapeutic strategies based on the CCR2 axis that block the monocyte/macrophage lineage in the treatment of cardiovascular disease models were successful in preventing cardiac injury.43,44 However, strategies that target monocyte/macrophages globally such as clodronate liposome administration therapeutically failed and in fact exacerbated cardiac injury.16 These results suggest that strategies based on the classically categorized M2 macrophage will need to be revisited and reviewed in the light of new subsets of M2 macrophages.

Figure 3

Four distinct subsets of cardiac macrophages under physiological and pathophysiological conditions. Ly-6Chi, Ly-6CloMHCIIhiCCR2CD11clo, Ly-6CloMHCIIhiCCR2+CD11chi, and Ly-6CloMHCIIlo macrophages are present in the normal mouse heart. Under conditions of cardiac stress, such as with angiotensin II infusion or myocardial infarction, the cell numbers of three of the four macrophage subsets (the Ly-6Chi, Ly-6CloMHCIIhiCCR2+CD11chi, and Ly-6CloMHCIIlo macrophages) are increased. On the other hand, the number of Ly-6CloMHChiCCR2CD11clo macrophages is decreased after cardiac stress. The Ly-6CloMHClo macrophage subset mainly expands after stress and contributes to the phagocytosis of debris of the heart. The Ly-6CloMHChiCCR2CD11clo macrophage possesses strong antigen-processing/-presenting function and contributes to immunosurveillance. On the other hand, the two CCR2-expressing subsets of macrophages, Ly-6Chi, and Ly-6CloMHCIIhiCCR2+CD11chi macrophages provoke cardiac inflammation. MHCII: major histocompatibility complex class II; CCR2: chemokine receptor 2; CX3CR1: CX3C chemokine receptor 1.

4. Phenotypic change of macrophages

In acute myocardial infarction, failure of transition from a M1 macrophage-dominant acute phase to a M2 macrophage-dominant late phase may induce prolonged M1 macrophage regulation and affect outcome of infarcted myocardium.14,38,39 Impaired resolution of inflammation pose as a major cause of various diseases.45 Thus, modulation of pro-inflammatory macrophages to anti-inflammatory macrophages is proposed as a new therapeutic approach.46 The pro-inflammatory macrophage may modulate its phenotype to become anti-inflammatory macrophages when phagocytosis of apoptotic cells occurs; in other words, M1 macrophages are able to modulate its phenotype into M2 macrophages with time.47 Apoptotic cells are recognized by macrophages via cell surface ligands, with phosphatidylserine (PS) being one such cell surface ligand. It is possible to phenotypically modulate M1 macrophages into anti-inflammatory macrophages by its phagocytosis of the PS-expressing liposome that mimics an apoptotic cell.47 Indeed, in vivo intravenous injection of PS liposomes into the post-infarcted rat showed that phagocytosis of PS liposomes by cardiac macrophages during the acute infarction phase reduced the expression of M1 markers, TNF-α and CD86, and increased expressions of the M2 marker, CD206, and anti-inflammatory cytokines, TGF-β and IL-10, indicating a M2 polarity shift.47 As a result, PS liposome-treated hearts had increased angiogenesis, reduced infarct area, and improved cardiac function compared with infarction controls.47 The study suggests that adequate adjustment of cardiac macrophage polarity and macrophage heterogeneity under pathological conditions can be cardioprotective, and failure of the adjustment in the polarity shift and timing results in deterioration of the disease.14,45

5. Mathematical analysis

The bivalent function of the cardiac macrophage can be analysed by physical chemistry law-based mathematical models and comprehensive genetic analyses of post-infarcted myocardium.48,49 These analyses do not consider pro-injury, apoptotic, and inflammatory M1 macrophages interfering with healing, anti-inflammatory, and pro-fibrotic M2 macrophages, but assume that both macrophages can interact and influence their respective genetic expressions over time to maintain homeostasis. When diseases disrupt the homeostasis, mechanisms of the homeostatic disruption can be investigated by large and comprehensive in silico data analyses. These comprehensive analyses, which do not focus on a single molecule, will be critical to characterize the complex and diverse properties of macrophages.48,49 These non-biased, analytical handling of big data is one of the novel methods that can be used to map the complex spatiotemporal biological processes in vivo.

6. Cardiac macrophages and hypertrophic signals

MicroRNA-155 is expressed in cardiac macrophages and T cells.5054 Since systemic knock-down of microRNA-155 in a myocarditis model can prevent infectious myocarditis, microRNA-155 within macrophages and inflammatory cells has been highlighted as a therapeutic target for myocarditis.51 In addition, cardiac hypertrophy and inflammation in a pressure-overloaded heart of macrophage-specific microRNA-155 knockout mouse were significantly prevented. The microRNA-155 induces cardiac hypertrophy and inflammation via SOCS1 gene regulation in macrophages, serving as a reminder that macrophages are also involved in cardiac hypertrophy.50

The early cardiac hypertrophic response to pressure overload is a compensatory and adaptive response. Indeed, mice with a lack of cardiac hypertrophic response exhibit early cardiac failure and die earlier.55 Many studies on pressure-overload cardiac hypertrophy using gene deletion mice or drug administration define their study endpoints in terms of progression or regression of cardiac hypertrophy. However, these endpoints are confounding as they can be interpreted as due to either a loss of pathogenic disease development or a loss of an adaptive response. Therefore, it may be better to define the study endpoints in terms of the development of cardiac failure or cardiac failure survival.

7. Macrophages as the therapeutic target in heart failure

It is well known that TNF-α secreted from macrophages suppresses cardiomyocyte contraction and myocardial contractility in vitro.56 Mechanical stretch induces TNF-α expression in cardiomyocytes, and the Tnf knockout mouse responds to pressure overload with milder ventricular remodelling, resulting in less cardiac hypertrophy and cardiac fibrosis.57 Furthermore, blood TNF-α levels in cardiac failure patients are increased,58 suggesting that TNF-α may be a cause in the onset of heart diseases.

The TNF-α secreted from cardiomyocytes binds to TNF-R on cardiac macrophages that then mediates further TNF-α secretion from cardiac macrophages. The secreted TNF-α may regulate other cytokines as a master regulator.5961 The TNF-α-mediated cardiomyocyte–macrophage interaction was targeted as ‘the axis of evil’ of cardiac failure, and a TNF-α receptor blocker, etanercept, and a chimeric monoclonal antibody to TNF-α, infliximab, were clinically tested as TNF-α suppression therapies. However, these trials did not improve heart failure and, in fact, worsened cardiac failure in some cases.62,63 Treatment with the synthetic guanylhydrazone, semapimod, which inhibits inflammatory cytokines including TNF-α, in a rat post-myocardial infarction model reduced TNF-α levels in the myocardium. However, cardiac function was worse and mortality rate was increased.64 The results of the animal experiment were consistent with those seen in the clinical trials.64 Thus, therapeutic strategies involving suppression of cytokines such as TNF-α are not effective for improving the outcome of cardiac failure patients.

The cardiomyocyte–macrophage interaction via TNF is likely complex. In mouse experiments of arteriosclerosis, Tnf  knockout mice bred with Apoe−/− mice had significantly reduced plaque formation,65 whereas Tnfra−/− mice had more plaque formation.66 In addition, TNF-α is also involved in macrophage differentiation.67 Therefore, it remains to be investigated how the cardiac macrophage maintains cardiac homeostasis and responds to stress conditions, and how disruption of the cardiac homeostasis triggers cardiac failure, by detailed analyses of macrophage heterogeneity and time-dependent changes of macrophages within the myocardium.

8. Perspectives

Cardioprotective effects of cardiac macrophages are important for the maintenance of cardiac homeostasis in various cardiac diseases, and disruption of the cardiac homeostasis may trigger onset of the cardiovascular diseases. However, the underlying mechanisms are not well understood yet. For instance, how can significantly small numbers of cardiac macrophages influence cardiomyocytes and cardiac fibroblasts within the myocardium; how can these different types of cardiac macrophages communicate with each other within the myocardium; what kind of stimulations can trigger cardiac macrophage induction/proliferation and when it happens; whether M1 and M2 macrophages are interchangeable or not; what is the boundary between M2 macrophages and dendritic cells; and whether malfunction of the cardiac macrophage can causes certain cardiac diseases, etc. It is obvious that further investigations are necessary.

Conflict of interest: none declared.


This study was supported by the ‘Funding Program for World-Leading Innovative R&D on Science and Technology (FIRST Program)’ (to R.N.), grants-in-aid for Scientific Research (S) and (B), and grants-in-aid for Young Scientists (B) from JSPS (23390203, 22229006, and 23790835) (to R.N. and K.F.); a grant for Translational Systems Biology and Medicine Initiative (to R.N.); and Precursory Research for Embryonic Science and Technology (PRESTO) (to K.F.) from JST.


  • This article is part of the Spotlight Issue on: Heterocellular signalling and crosstalk in the heart in ischaemia and heart failure.


View Abstract