Monocytes and macrophages in cardiac injury and repair

Abstract

Myocardial infarction (MI) is one of the major contributors to worldwide morbidity and mortality. It is atherosclerosis’ most dreadful complication and occurs after the supply of oxygenated blood to the heart is blocked. Understanding how cardiac tissue is injured and later regenerates is of crucial importance to improve the sequelae after the acute event. We now understand that the immune system substantially contributes to both the acute inflammatory response and the regenerative response that follow tissue injury after MI. In this review, we will focus on the role of monocytes and macrophages, which are cellular protagonists of the immune system, in acute cardiac injury and post-MI repair.

Monocytes/macrophages during cardiac homeostasis

In the steady state the heart is populated with cardiac resident macrophages (Figure 1A) (1-3). These cells exhibit a spindle-like shape and are interspersed between cardiomyocytes (4). Under physiological and pathological conditions, cardiac resident macrophages appear not to be a homogenous population, but rather comprise distinct subsets, depending on expression levels of Ly6C, CCR2, CD11c, and histocompatibility-2 (5). In mice, these macrophages populate the heart during embryogenesis and originate primarily from yolk sac progenitors. Once residing in the heart, these macrophages self-maintain through local proliferation and seem to be independent of bone marrow (BM) derived blood monocytes (3,5). However, it has been proposed that blood monocyte-derived macrophages also seed the heart and gradually replace cardiac resident macrophages (6). The function of cardiac resident macrophages during cardiac homeostasis is less well investigated and may include phagocytosis of aged or dying cardiomyocytes, angiogenesis, fibrosis, and immunosurveillance (antigen processing and presenting) (7).

Figure 1.

Monocytes/macrophages in cardiac homeostasis and during injury and repair. (A) Cardiac resident macrophages (red circles) populate the heart during steady state. They seem to self-maintain through in situ proliferation and are mostly independent from blood monocytes (white circles); (B) after acute ischemia (MI), cardiac resident macrophages disappear and huge numbers of blood Ly6C^high monocytes are recruited to the ischemic area; (C) in mice with ischemic cardiomyopathy (HFrEF), macrophage numbers in the scar wane, while numbers gradually increase in the remote myocardium. This remote macrophage expansion is caused by both, increased local proliferation as well as increased blood monocyte recruitment. MI, myocardial infarction.

Monocytes/macrophages during acute myocardial infarction (MI)

MI, the most dreadful complication of atherosclerosis, occurs after atherosclerotic plaque rupture/erosion resulting in atherothrombosis which blocks the supply of oxygenated blood to the heart. Resident macrophages from the ischemic area disappear, probably because they become necrotic and/or apoptotic (Figure 1B) (8,9). MI inflicts a sterile wound on the heart that recruits hundreds of thousands of leukocytes from the blood to the ischemic heart (8). After a neutrophil peak, inflammatory blood Ly6C^high monocytes are recruited and accumulate in the infarct area. Recruited Ly6C^high monocytes give rise to inflammatory macrophages, which both remove dead cells and extracellular matrix debris by phagocytosis and secret proteases and reactive oxygen species.

Blood monocytes contribute to up to 10% of the total blood leukocyte counts in both mice and humans (2) and are a heterogeneous population (10). There are two subsets of circulating CD115⁺ monocytes in mice, Ly6C^high (CCR2⁺CD62L⁺CX3CR1^mid) and Ly6C^low (CCR2⁻CD62L⁻CX3CR1^hi) monocytes.

The inflammatory Ly6C^high monocytes are recruited to the heart following ischemic injury at a high rate, produce high levels of proinflammatory cytokines [such as tumor necrosis factor-α (TNFα) and interleukin-1β (IL-1β)], and differentiate into inflammatory macrophages after extravasation (11). Recruitment of Ly6C^high monocytes relies on the interaction of monocytic CCR2 and endothelial chemokine (C-C motif) ligand 2 (CCL2)/chemokine (C-C motif) ligand 7 (CCL7) (12). Apart from these chemokines, cell adhesion molecules expressed on both, blood monocytes and cardiac endothelial cells, also contribute to enhanced leukocyte recruitment. It was reported that intercellular adhesion molecule 1 (ICAM1), vascular cell adhesion protein 1 (VCAM1), E- and P-selectin are upregulated in the remote myocardium (the non-ischemic part) in mice with heart failure with reduced ejection fraction (HFrEF) (3) and that simultaneous blocking of endothelial ICAM1, ICAM2, VCAM1, E- and P-selectin reduced monocyte recruitment in the setting of acute MI (13).

The Ly6C^low monocytes patrol the vascular lumen, adhere to and move along the endothelium, both clearing damaged cells and triggering inflammatory responses (14-16). In the blood, Ly6C^high monocytes can convert to Ly6C^low monocytes (17). Whether Ly6C^low monocytes are recruited to the heart and contribute to the cardiac macrophage pool is currently discussed (8,18).

The human blood harbors three subsets of monocytes: Classical inflammatory monocytes (CD14⁺CD16⁻, resembling murine Ly6C^high monocytes), non-classical monocytes (CD14^lowCD16⁺ monocytes, resembling murine Ly6C^low monocytes), and an intermediate monocyte subset (CD14⁺⁺CD16⁺CCR2⁺CX3CR1⁺⁺CCR5⁺) (19).

Splenic monocyte release

The infarct recruits hundreds of thousands of leukocytes per day, a process that depletes the blood pool quickly. In order to meet this high demand, leukocytes are released from emergency reservoirs. In mice, the spleen functions as such a monocyte reservoir and contributes ~50% of myeloid cells that reside in the infarct in the early hours after coronary ligation (20). Monocyte release from the red pulp of the spleen thereby depends on angiotensin II/angiotensin 1 receptor interplay. Angiotensin II administration mimicked the release of splenic monocytes into the blood as observed after MI (21). Moreover, splenectomy leads to adverse infarct healing indicating that splenic monocytes may have a protective role (22). Accordingly, administration of the angiotensin-converting-enzyme inhibitor enalapril early after MI did not improve survival (23).

Splenic monocyte production

Apart from releasing leukocytes in the setting of MI (20), the spleen can function as a leukocyte production site (extramedullary hematopoiesis) especially during later stages of MI (24). The MI activates the sympathetic nervous system of the BM which releases its neurotransmitter norepinephrine. Norepinephrine signals to BM parenchymal cells that lower the expression levels of Cxcl12, Scf, and Vcam1. This in turn results in release of hematopoietic stem and progenitor cells into the blood. These cells seed the spleen and induce extramedullary hematopoiesis (24).

BM monocyte production

The increased demand for myeloid cells also fuels emergency hematopoiesis. Monocytes are mononuclear phagocytes that are generated at medullary or extramedullary sites in a process termed myelopoiesis. Granulocyte/macrophage precursors (GMPs) originate from hematopoietic stem and progenitor cells and mature into macrophage (monocyte)/dendritic cell precursors (MDPs), which give rise to Ly6C^high monocytes through common monocyte progenitors (cMoPs) (25).

The role of the BM in the inflammatory response that follows MI was recently investigated (26). BM hematopoietic stem cells (HSC) are activated after MI and give rise to a subset of CCR2+ HSCs that proliferate vigorously in response to ischemia increasing the production of myeloid cells (monocytes, macrophages, neutrophils). These myeloid cells are deployed to the cardiac wound and promote infarct healing. In patients with acute MI, a similar BM activation was observed after ischemia (26).

Further, the signals that drive this BM activation after MI were studied (27). A crosstalk of heart and BM is established after MI: the pro-inflammatory cytokine IL-1β is released from the ischemic heart, signals to the BM and drives HSC proliferation leading to an increase in production of myeloid cells. BM reconstitution experiments revealed that this increase in HSC proliferation occurs through a direct action of IL-1β on HSC, but also indirectly by altering the HSC microenvironment (HSC niche). Here, niche cells lower the expression of HSC maintaining factors and hence trigger proliferation and release of HSC. Anti-IL-1β treatment dampened BM activation, decreased production of myeloid cells and ultimately lowered the content of inflammatory cells in the infarcted tissue. In turn, this resulted in a better post-MI remodeling (27).

BM monocyte release

Monocyte release from the BM depends on CCL2/CCR2 and CCL7/CCR2 signaling (28,29). Upon stimulation, BM stromal cells produce CCL2 that can be sensed by CCR2⁺ monocytes, which then egress out of the BM into the circulation (30). Genetically modified mice lacking CCR2 do not only show an impaired recruitment of leukocytes to sites of inflammation, they also exhibit a pronounced blood monocytopenia because their monocytes are retained in the BM (29). Furthermore, CXCL1/CXCR2 signaling also contributes to blood monocytosis. In mice fed a high-cholesterol diet, CXCL1 neutralization dampens expansion of blood Ly6C^high monocytes (31).

Monocytes/macrophages in the post-MI phase

The early inflammatory phase is then succeeded by a reparative phase (starting on day 3 after MI) (32). Here, macrophages functionally shift from promoting inflammation to supporting repair and produce growth factors that recruit and activate mesenchymal reparative cells (mainly myofibroblasts and vascular cells) (33). This transitioning from inflammatory to reparative macrophages resembles in vitro polarization from the so-called “M1” to “M2” macrophages. After ingesting dead cells (a process called efferocytosis) macrophages decrease production of pro-inflammatory cytokines, such as IL-1β and TNFα, and increase production of anti-inflammatory and pro-fibrotic cytokines, such as IL-10 and transforming growth factor-β (TGFβ) (34,35). This indicates that efferocytosis may cause a shift in macrophage activation toward M2-like cells (36). However, macrophages display a high plasticity and can rapidly adapt their phenotype to different stimuli. Hence, the classic M1/M2 macrophage characterization, which originates from in vitro studies, most likely does not reflect phenotypic changes macrophage undergo in the infarct and needs careful re-evaluation (37).

The question arises whether macrophage phenotype/function is solely dictated by the local environment that surrounds these cells or also depends on their origin/ontogeny (resident vs. blood monocyte derived macrophages) (32). A recent study using fate mapping experiments revealed that macrophages from different origins in heart failure mice also display distinct phenotypes (3). Although it was not evaluated whether macrophages from varying sources reside in distinct locations in the heart, and are consequently embedded in different environments.

Ly6C^low macrophages facilitate wound healing and regeneration by (I) promoting transdifferentiation of fibroblasts into secretory and contractile myofibroblast (α-smooth muscle actin-positive fibroblast that are the major source of collagen for extracellular matrix deposition) through TGFβ secretion; (II) promoting angiogenesis through VEGF secretion (33); and (III) altering the extracellular matrix built-up/break-down through regulating the balance of matrix metalloproteinases (MMPs) and their inhibitors (tissue inhibitors of MMPs). Subsequently, the infarct matures and a collagen-based scar is formed. Depletion of monocytes/macrophages impairs scar formation by attenuating collagen deposition and angiogenesis (33).

Although the inflammatory phase is essential for removal of dead cell and debris, a timely repression of the inflammatory response is critical for effective healing. Exacerbated and prolonged inflammation may impede the healing process, as it was shown in infarcted mice with atherosclerosis that displayed adverse cardiac remodeling and blood monocytosis (38). Accordingly, elevated blood monocyte counts in patients after MI negatively correlate with ejection fraction (39). Along these lines, humans and mice with heart failure also display a blood leukocytosis (3,40,41). In general, elevated white blood cell numbers are an independent risk factor for CAD (42-44) and more specifically, higher monocyte counts independently predict risk for atherosclerosis after adjustment for conventional risk factors (45-49).

Macrophages in post-MI heart failure

Ventricular remodeling (defined as alterations in size, shape, structure and physiology) of both the infarcted and non-infarcted (remote) myocardial segments is closely linked with the development of post-MI heart failure (4). Recently, the involvement of monocytes/macrophages in the development and progression of ischemic cardiomyopathy was described (3,50). It was reported that macrophages gradually expand within the remote myocardium during the course of heart failure, while macrophage numbers within the scar wane (Figure 1C) (3). This remote macrophage expansion is caused by (I) increased local proliferation and (II) increased blood monocyte recruitment. It was reported that mechanical strain can stimulate macrophage proliferation via activating the MAPK pathway, indicating that higher mechanical strain in the failing heart may promote local macrophage proliferation. Furthermore, in HFrEF mice an increased sympathetic tone elevates systemic myeloid cell production in the BM causing the observed blood leukocytosis. Recruited monocytes contribute about one third to the increase in total cardiac macrophages. Using RNAi nanoparticle technology, the authors reduced post-MI monocyte recruitment to the remote myocardium, which attenuated adverse left ventricular remodeling. These data provide evidence that inflammatory BM derived cardiac macrophages augment adverse post-MI heart remodeling.

Conclusions

The post-MI phase is characterized by an early inflammatory stage where dead cells and debris are cleared from the infarct area and a subsequent reparative stage where fibrosis and angiogenesis occurs. Recent findings revealed the contribution of immune cells to cardiac injury and repair. Monocyte/macrophages are highly plastic cells that can rapidly adapt their phenotype/function and hence contribute differently to each stage of the disease. During the inflammatory phase they phagocytose necrotic cells and fuel inflammation by releasing proteases, pro-inflammatory cytokines/chemokines, and reactive oxygen species. In the reparative phase macrophages undergo transformation and exert pro-fibrotic functions by promoting collagen production and neovascularization. A timely resolution of each phase is crucial since an overzealous and prolonged inflammation and/or fibrosis may be deleterious and determines whether progression to heart failure or repair of cardiac tissue occurs. These experimental observations may pave the way for clinical studies that aim to target this biphasic response to improve outcomes following MI.