Cardiac Macrophage Metabolism in Health and Disease

Abstract

Cardiac macrophages are essential mediators of cardiac development, tissue homeostasis, and response to injury. Cell intrinsic shifts in metabolism and availability to metabolites regulate macrophage function. The human and mouse heart contain a heterogeneous compilation of cardiac macrophages that are derived from at least 2 distinct lineages. In this review, we detail the unique functional roles and metabolic profiles of tissue resident and monocyte-derived cardiac macrophages during embryonic development, adult tissue homeostasis, and in response to pathologic and physiologic stressors. We discuss the metabolic preferences of each macrophage lineage and how metabolism influences monocyte fate specification. Finally, we highlight the contribution of cardiac macrophages and derived metabolites on cell-cell communication, metabolic health, and disease pathogenesis.

Macrophages are a Dominant Cell Population within the Heart

The heart is responsible for pumping blood throughout the body, contracting more than 2.5 billion times over an average lifespan. With each heartbeat, cardiomyocytes undergo synchronized and energetically demanding contraction-relaxation cycles with a daily ATP metabolic demand 20 times the heart’s mass [1]. While an extensive body of work has been dedicated towards understanding cardiomyocyte metabolism (in part because cardiomyocytes represent the majority of cellular mass of the myocardium) [2], it is important to recognize that cardiomyocytes comprise only 30% of the total number of cells within the heart. The remaining 70% of cardiac cells include immune cells, endothelial cells, fibroblasts, pericytes, and smooth muscle cells and are essential for proper cardiomyocyte function. In recent years, metabolic pathways amongst these cardiac cell populations have been increasingly studied [3, 4]. Cardiac macrophages represent 6–10% of all cardiac cells, are the most abundant immune cell in the heart [5, 6], and are key modulators of cardiac health and disease [7]. In this review, we will discuss the metabolic preferences of cardiac macrophages across stages of life and in response to physiologic (aerobic exercise) and pathologic (myocardial infarction, diabetes) perturbations. We will focus on the role that macrophage metabolism plays on macrophage differentiation, function, and signaling. Over the past several years, implementation of new and powerful technologies has provided new insights into the vast extent of macrophage diversity. The rapid adoption of lineage tracing and single cell RNA sequencing coupled with advancement in our understands of macrophage metabolism have uncovered new avenues and opportunities for investigation and further invigorated the growing fields of cardio-immunology and immunometabolism.

Macrophage Diversity in the Heart

The functional diversity of both recruited and resident macrophages extends well beyond the initial M1/M2 classification paradigm. While it has been understood for some time that macrophages display a wide range of behaviors and phenotypes, only recently have the tools emerged (genetic lineage tracing and parabiosis) to dissect the ontologic diversity of macrophages in vivo [8, 9] and ascertain their functional roles in health and disease [10]. Cardiac resident macrophages (CRMs, see Glossary) are derived from embryonic hematopoietic progenitors that reside within the yolk sac and fetal liver, seed the heart during mid-gestation, and are maintained throughout life independent of monocyte input. CRMs seed the heart during embryonic development based on instructive cues from the epicardium [11] and expand through local proliferation. CRMs lack the cell surface expression of C-C chemokine receptor 2 (CCR2⁻) and instead express a core tissue resident macrophage signature consisting of CD163, LYVE1, FOLR2, F13A, TIMD4, and CD169. CRMs lack inflammatory potential and instead express transcriptional signatures associated with wound healing and angiogenesis [9, 12, 13]. Monocyte-derived macrophages (MDMs) originate from definitive hematopoietic stem cells within the bone marrow and spleen and traffic to the heart during times of injury and stress through classical monocyte intermediates. Classical monocytes and most MDMs express CCR2 (CCR2⁺), require continuous replenishment over time, and accumulate in the context of aging and disease. Human and murine markers generally overlap (Table 1). Genetic fate mapping studies have shown that across organs (heart, liver, lung, kidney, brain) tissue resident macrophages can be generally identified by a shared subset of cell surface markers (CD163, TIMD4, LYVE1, FOLR2) whereas monocyte-derived macrophages are generally marked by the expression of CCR2 [14]. CRM and MDM are admixed within the heart and display differential responses to the same microenvironment. Recent single cell RNA sequencing studies have suggested that MDMs in the mouse and human heart are exceedingly diverse and represent a compilation of multiple transcriptionally distinct subsets [14–17] controlled in part by epigenetic regulation [18]. MDM can be quite diverse and dynamic and are generally considered to have robust inflammatory potential; however, the exact function of each sub-population is an area of active exploration [16].

Table 1:

Murine and Human Macrophages can be defined by their expression of CCR2 [14, 19].

	Mouse	Human
Pan Macrophage	F4/80, MHC, CD64, CD68, MAC-3, GAL-3	CD14, CD16, MHC, CD64, CD68, EMR1, MAC-3, GAL-3
CRM	CCR2−, FOLR2, F13A, TIMD4+ or −, CD169, CD163, LYVE1+ or −, TLF+ or −, CX3CR1hi or lo, MERTK	CCR2−, CD163, LYVE1, MERTK
MDM	CCR2+, MHChi; CX3CR1hi	CCR2 + ; HLA-DR

Cardiac Macrophage Functions during Fetal and Neonatal Development

During early development, CRMs are essential for vascular development and heart regeneration. CRMs reside next to the coronary vasculature and are essential for coronary vascular development, growth, and maturation potentially through insulin-like growth factor signaling [20]. CRMs also regulate neonatal heart regeneration (an ability that is lost shortly after birth) possibly through an enhanced capacity to clear dysfunctional cardiomyocytes through efferocytosis [21–23]. Conversely, MDMs populate the heart later during embryonic and postnatal development [9, 24] and initially reside adjacent to the trabecular projections of the endocardium [20]. While the precise function of MDMs in cardiac development and early postnatal life is incompletely understood, their pro-inflammatory gene signature and ability to be activated by damage associated molecular patterns (DAMPs) though toll-like receptors (TLRs) suggest they may coordinate responses to injury or infection [8, 9] (Table 2).

Table 2:

CRMs and MDMs have Unique Origins and Developmental Functions

	Cardiac Resident Macrophage	Monocyte Derived Macrophage
Origin	Embryonic hematopoietic progenitor from yolk sac and fetal liver	Definitive hematopoietic stem cell from bone marrow
Enter the Heart	Mid-gestation	After CRMs/during times of stress/with aging
Location within Embryonic Heart	Reside next to coronary vasculature	Trabecular projections of the endocardium
Expansion	Local proliferation	Continuous replenishment with classical monocytes
Developmental Function	Vascular development/heart regeneration	Unclear, possibly injury/infection response
Gene Expression	Anti-inflammatory	Pro-inflammatory

Cardiac Macrophage Metabolism during Fetal and Neonatal Development

Macrophages have distinct metabolic requirements and metabolic profiles at different stages of development in part influenced by the local and global metabolic environments [25, 26]. These metabolic shifts within cardiac macrophages determine macrophage function which in turn help to maintain tissue level and systemic metabolic homeostasis [27–30]. The heart develops within the hypoxic in utero environment where glycolysis and lactate metabolism serve as the primary pathways for the generation of ATP. As such, glucose, lactate, and pyruvate are the primary fuels in the fetal heart. While perturbations to substrate availability (such as maternal hyperglycemia) are well known to impact cardiomyocyte development [31], less well-defined is how similar metabolic derangements impact cardiac macrophage maturation and function. Metabolic tracing of ¹³C pyruvate through magnetic resonance imaging can be used to assess in vivo macrophage metabolism and to impute activation of the macrophages [32, 33], as ¹³C pyruvate is converted to ¹³C lactate in activated macrophages. Fetal macrophages utilize lactate to fuel oxidative phosphorylation through acetyl CoA. In addition to being a fuel source for oxidative metabolism within macrophages, lactate may serve as a signaling mediator as a ligand for G-coupled protein receptor 81[34]. Lactate can rewire macrophages to behave in an immunomodulatory or tolerogenic manner by reducing NF-κB activation [35]. Over time, macrophages are exposed to maternal nutrition [23]. Mononuclear cells isolated from infants fed breast milk (containing high levels of polyunsaturated fatty acids (PUFAs) including omega-6 arachidonic acid) or formula supplemented with PUFAs expressed higher levels of IL-10, which was linked with fetal healing [36]. Maternal nutrition is also important to regulating the expression of peroxisome proliferator- activated receptor (PPAR) gamma [37]; neonatal macrophage PPARγ activity is associated with increased fatty acid catabolism and is sufficient to activate endothelial progenitor cells to induce angiogenesis and myogenesis [38]. Within the muscle stem cell niche, anti-inflammatory macrophages express PPARγ, which promotes muscle regeneration through growth differentiation factor 3 [39] and through “trans-repressing” inflammatory factors such as NF-κB [40]. In addition to maternal metabolites, the neonatal heart is subjected to low oxygen availability (1%–5%) in utero which favors macrophage expression of hypoxia inducible factors (HIFs) as evidenced by fetal macrophages expressing higher levels of HIF1α than their adult counterparts [41]. HIFs help to coordinate a transcriptional program that shifts cellular metabolism to prioritize glycolysis and are critical to the production of eicosanoids, bioactive lipids that regulate tissue inflammation and repair like prostaglandins, thromboxanes, PUFAs [23, 42, 43], through HIF dependent transcription of cyclooxygenase 2 [44].

At birth, there is an uncoupling of the proton gradient in endotherms necessary to produce internal heat resulting in an increase in reactive oxygen species (ROS). While it is known that ROS drive cardiomyocyte cell cycle exit through DNA damage [23], the impact on macrophage maturation is less well understood. Oxidative metabolism of lactate results in generation of hydrogen peroxide, which is suggested to contribute to cardiac regeneration [45]. Additionally, neonatal macrophages transiently upregulate Ccl24 and Clcf1 and their expression coincides with the brief window of cardiac regenerative capacity [46]. Further work is necessary to understand whether the expression of Ccl24 and Clcf1 during this regenerative window represents a coordination between metabolism, cardiac macrophage growth factor expression, and regenerative potential.

To this point, we have spoken generally about neonatal macrophage metabolism. It remains unknown if CRMs and MDMs display differential shifts in metabolism during each stage of heart development. Consistent with a role in in coordinating cardiac regeneration, CRMs express HIF1α [23, 47, 48]. However, HIFs are also induced on MDMs through Toll-like receptor (TLR) signaling in the setting of myocardial infarction (through disruption of the TCA cycle) [49, 50], further work is necessary to deconvolute CRM and MDM metabolism during development.

Shortly after birth, the healthy adult heart reprioritizes its fuel sources with a stronger dependency on fatty acids (60–70%) rather than glucose (30–40%) as its major fuel source [51]. Along with this shifting fuel source, there is growing evidence of metabolic cross-talk between macrophages and other major cell types in the heart [23] (Figure 1) in maintaining tissue homeostasis, adapting to physiobiological changes, and the pathogenesis of cardiac remodeling and disease pathogenesis.

Figure 1: Metabolic cross-talk between macrophages and stromal cells within the heart.

There is growing evidence that within the heart as well as in other tissues that macrophages and stromal cells interact within their shared microenvironment through metabolic signaling. Some examples are shown. For instance, lactate has been shown to induce dedifferentiation whereas glutamine is important in cardiomyocyte proliferation [52]. Succinate can bind to cardiomyocyte G-protein coupled receptor 91 to cause hypertrophy [53]. Urokinase plasminogen activator (uPA) activates adipose secreted PDGF-D into active PDGF-DD leading to adverse cardiac remodeling in AngII infused obese mice [54]. Macrophage derived maresin1 (a specialized pro-resolution mediator (SPM) activates LGR6 to inhibit smooth muscle activation [55]. Within the fibroblast compartment, the SPM MaR1 is thought to function through Nrf2 to reduce fibroblast activation and improve cardiac fibrosis [56, 57]. Macrophages interact with endothelial cells through numerous mediators, including IGF-1, VEGF, and MaR1 [58–61]. Recently, the secretome of the pericyte is being unraveled and it is becoming clear that the pericyte is a key mediator of cardiac homeostasis [62, 63]. LTC4: Leukotriene C4. Figure created with BioRender.com.

Role of Macrophages in Maintaining Tissue Health

As post-mitotic cardiomyocytes have limited capacity to self-renew, maintaining cardiomyocyte integrity is essential to health. A recent study uncovered that cardiomyocytes eject damaged or dysfunctional mitochondria through cardiac exophers, which are captured and eliminated by cardiac macrophages, predominately CRMs (binding to MerTK). Mitochondrial proteins and DNA are damage associated molecular patters (DAMPs) and serve as intracellular and extracellular triggers of inflammation [64]. The ability of macrophages to clear ejected cardiomyocyte mitochondria (heterophagy) [26, 65] is essential to maintaining cardiac function with heterophagy responsible for recycling of up to 60% of mitochondria in the heart. Depletion of macrophages leads to accumulation of mitochondria and cardiomyocyte contractile dysfunction [65].

In addition to recycling cardiomyocyte mitochondria, macrophages have other salient and homeostatic functions. Macrophages clear dying cells from the heart and recycle their metabolites [26]. Additionally, macrophages release extracellular vesicles loaded with microRNAs, some of which, modulate glucose and insulin sensitivity in cardiomyocytes [66, 67].

Macrophages respond to many other metabolites including iron, calcium, and amino acids. For example, macrophages scavenge iron and intracellular iron content contributes to the progression of atherosclerotic plaques [68]. Iron metabolism is also essential for ferroptosis, a form of inflammatory programmed cell death that is dependent on iron. Intriguing studies by the Kagan group have shown that macrophage subsets have differential susceptibility to ferroptosis [69]. Macrophages may also transfer mitochondria to induce ferroptosis in cardiomyocytes [70] highlighting the possibility that macrophages may be important mediators of cardiac cell death. Similarly, macrophages uptake calcium in atherosclerosis [71] where TRPV4 channels are important for macrophage calcium homeostasis and function [72]. Recent studies have established the importance of calcium signaling between macrophages and cardiomyocytes in maintaining cardiac electrical activity [73, 74]. Amino acid have also been shown to have divergent roles in the activation and function of macrophages. For example, arginine has been implicated to suppress IL-6 and IL-1β production [75]; however, when L-arginine was given post-MI, outcomes were worse, possibly due to arginine metabolism to NO which can interfere with mitochondrial metabolism and the TCA cycle [7, 76, 77]. Together, these findings exemplify the impact different metabolites may have on macrophage function within the heart.

Macrophage Metabolism in Inflammation and Repair

The response to ischemic myocardial injury involves a coordinated activation and signaling amongst immune and cardiac stromal cells [74, 78]. This begins with an early inflammatory response (inflammatory phase) balanced by a later tissue healing program that resolves inflammation (resolution phase) to prevent maladaptive and excessive scaring. Central to this construct is tight regulation of macrophage metabolism. Diverse Role of Macrophages during the Inflammatory Phase after Myocardial Infarction In the setting of myocardial ischemia, cardiomyocytes may become damaged. Macrophages have been shown to minimize tissue injury through cardiomyocyte clearance. The tissue ischemia and increased phagocytic burden of myocardial ischemia pose significant metabolic stresses on macrophages and drives a shift in macrophage metabolism to modulate the inflammatory and tissue repair responses. During the initial periods of ischemia, hypoxia and nutrient deprivation drive an increase in glycolysis and lactate generation in macrophages through the nutrient sensor AMPK [79, 80]. This is a conserved response that is also evident in skeletal macrophages, where AMPK is upregulated and is necessary for macrophage oxidative phosphorylation. AMPK-deficient macrophages are unable to trigger myogenesis despite IL-10 or IL-4 exposure [81]. Hypoxia sensing in macrophages is mediated through HIF1 signaling which prioritizes glycolysis and lactate production [43, 82]. Additively, DAMPs released from damaged cardiac tissue can propagate HIF1α through TLR signaling.

After myocardial infarction (MI), there is an increase in exopher production in viable myocardial regions adjacent to the infarct zone. In addition, apoptotic cells are efferocytosed by macrophages [79]. MerTK is expressed on the surface of macrophages and recognize phosphatidylserine on the surface of exophers and apoptotic cells [83]. Efferocytosis directly impacts metabolism, monocyte and macrophage polarization, and triggers the release of mediators [79] that facilitate inflammation resolution tissue repair [84]. Consistent with a role of MerTK expression and resolution, cleavage of MerTK worsened remodeling after MI [21].

This is contrasted by the macrophage upregulation of AXL that uniquely promotes a pro-inflammatory state through a metabolic switch that favors glycolysis [85]. Consistent with a pro-inflammatory states dependence on glycolysis, inhibition of glycolysis with 2-deoxyglucose attenuates inflammation [4]. MDMs are hyperinflammatory and initially rely on glycolytic metabolism and utilize the pentose phosphate pathway (PPP) to generate NADPH to synthesize pro-inflammatory lipid mediators [3]. Increased glycolysis and PPP lead to an increase in pyruvate, including the glycolytic enzyme pyruvate kinase muscle enzyme 2 (PKM2) which promotes the transcription of IL-6 and IL-1β [86]. PKM2 interacts with HIF1α to enhance inflammation and the loss of HIF1α in monocytes reduces inflammation [87, 88]. An increase in pyruvate coupled with an inefficient TCA cycle leads to an accumulation of succinate which stabilizes HIF1α and further promotes pro-inflammatory signaling through IL-1β [3, 89].

MDMs secrete and sense extracellular succinate through the G-protein receptor 91 (GPR91) [90]; thus, acting in an autocrine or paracrine manner. Succinate is then oxidized leading to mitochondrial ROS, further activating HIF1α [91]. GPR91 is also expressed on and regulates calcium within the cardiomyocyte leading to cell death or pathologic hypertrophy [53] (see earlier discussion on calcium signaling). MDM activation leads to other TCA intermediates aside from succinate, including itaconate. Interestingly, as opposed to pro-inflammatory succinate, itaconate has anti-inflammatory properties [92] suggesting that MDMs have some degree of feedback inhibition through modulation of succinate dehydrogenase (SDH), as succinate is metabolized through SDH whereas itaconate antagonizes SDH [89] (Figure 2).

Figure 2: Metabolic differences between macrophage populations help define their function.

Early in the inflammatory phase, pro-inflammatory macrophages utilize glucose and succinate in the hypoxic microenvironment. Pro-inflammatory macrophages use these metabolites to prioritize glycolysis and the PPP. These metabolic changes help to enhance macrophage efferocytosis, amplify the inflammatory response, recruit inflammatory cells, and facilitate interaction with stromal cells. With resolution, macrophages prioritize different fuel sources to prioritize oxidative phosphorylation and increase their utilization of fatty acids. This change in fuel aids in the suppression of inflammation and potentiates wound healing. Figure created with BioRender.com.

It is becoming more evident that the metabolic reprogramming that occurs during the inflammation and resolution phases is modulated by mitochondrial function. Indeed, macrophage specific deletion of the mitochondrial protein (Ndufs4) impairs efferocytosis [93]. Efferocytosis increases fatty acids within macrophages and is facilitates metabolic signaling through NAD⁺[94]. Interestingly, the anti-inflammatory IL-10 signal is lost in macrophages that have a mitochondrial complex III defect which can be rescued by adding NAD⁺ precursors, which activate SIRT1 and subsequently the IL-10 transcription factor PBX1 [94]. Metabolic responses observed in macrophages are also evident in cardiac stromal cells, further supporting the immune-stromal metabolic crosstalk and the importance of considering the local environment rather than individual cells. Fibroblast metabolism is mediated by HIF1α [95], which inhibits pyruvate entry into the mitochondria. Activated fibroblasts rely on glucose and α-ketoglutarate (αKG) to drive glycolysis, resulting in the expression of pro-inflammatory and pro-fibrotic gene profiles [51]. Interestingly, endothelial cells rely on glycolysis under steady state conditions and actively shuttle fatty acids to cardiomyocytes. Following acute cardiac injury, endothelial cells continue to prioritize glycolysis to support cardiomyocyte survival. However, in the context of chronic injury, glycolysis becomes impaired and metabolism shifts towards oxidative phosphorylation, contributing to endothelial cell dysfunction [51]. Lastly, cardiac macrophages modulate T-cell trafficking and expansion, highlighting the role of CRMs and MDMs as key upstream regulators of adaptive immune responses [96]. Splenic CD14⁺ CD16⁺ monocytes recruited post MI have been shown to be additional drivers of inflammation [97]. While the recruitment of splenic monocytes is dependent on CCR2-CCL2 signaling, mechanisms of monocyte trafficking to and within the myocardium is complex. Endothelial cell capture, rolling, crawling, and transmigration depend on a host of monocyte and endothelial adhesion molecules, inflammatory cytokines/chemokines, and metabolic states [98, 99].

Macrophages are Essential to the Resolution Phase after Myocardial Infarction

Coincident with inflammatory resolution, macrophages within the heart prioritize oxidative forms of metabolism and de-prioritize glycolysis and the PPP. This leads to a number of important changes including increases in: 1) αKG to succinate ratio [100], 2) fatty acid uptake and utilization [101], 3) production of resolvin and other specialized pro-resolution lipid mediators (SPMs) from docosahexaenoic (DHA) and other PUFA substrates, 4) IL-10 production and transcriptional repression of NF-κB expression [3, 100] mediated through restoration of NADH levels [102–105], 5) production of anti-inflammatory arachidonic acid metabolites including, lipoxins prostaglandins, prostacyclins, and thromboxanes that promote stromal and vascular smooth muscle cell proliferation through the HIPPO/YAP pathway [106], and 6) glutamine synthetase activity.

The above metabolic changes in macrophages represent important opportunities to suppress inflammation and potentiate wound healing. NAD restoration leads to increases in αKG, which aids in HIF1α degradation and anti-inflammatory macrophage polarization [100, 107]. Direct targeting of metabolic switches may also be of benefit: treatment of mice with an aspartate aminotransferase inhibitor suppresses macrophage glycolysis, which polarizes macrophages toward an anti-inflammatory phenotype and reduces infarct size [108]. Conversely, inhibition of branched chain amino acid transaminase 1 is sufficient to reduce macrophage inflammatory behavior [109]. A third pathway of inflammation resolution has focused on resolvins and other SPMs [110–113]. There are numerous examples of powerful effects of SPMs; for instance, Maresin1 inhibits the proliferation and differentiation of fibroblasts into myofibroblast [114].

It remains unclear as to whether metabolic adaptations have different effects on CRMs and MDMs. Recruitment of MDMs drives the inflammatory response after MI [13] while CRMs suppress inflammation and trigger reparative events including coronary angiogenesis [72]. CRMs and MDMs are exposed to different metabolic stimuli. CRMs are lost from the infarct region and replaced by infiltrating monocytes and MDMs through rapid monocyte recruitment and subsequent proliferation [115]. CRMs predominate within regions of the myocardium remote to the infarct whereas the border zone contains an admixture of CRMs and MDMs. It is likely that monocytes and MDMs drive the initial inflammatory response and are activated within the hypoxic and inflammatory milieu of the infarct zone. Within the border zone, CRMs and MDMs are exposed to cytokines and reactive oxygen intermediates that may differentially impact metabolism and cell signaling. The remote zone experiences mechanical strain that increases metabolic demand. To date, the relative balance between glycolysis, fatty acid oxidation, and ketone utilization and their physiological consequences in CRMs and MDMs within each of these myocardial niches remains incompletely understood. However, the field is now poised to address these important questions using emerging tools and technologies to precisely target CRMs versus MDMs and measure spatially restricted gene expression and metabolism [1, 116–118].

Cardio-metabolic Heart Disease

While ischemic heart disease often leads to heart failure with reduced ejection fraction (HFrEF), there is a growing prevalence of heart failure with preserved ejection fraction (HFpEF) which often develops from extracardiac stressors, including obesity and diabetes [119–121]. HFpEF is largely driven by systemic metabolic abnormalities with additional contribution from ischemia.

Circulating fatty acids, ceramides, and glucose promote pro-inflammatory macrophages leading to cardiac fibrosis and diastolic dysfunction [122, 123]. Impaired insulin signaling and mitochondrial dysfunction are two important mechanisms implicated in the development and progression of diabetic cardiomyopathy [124]. Abnormal insulin signaling leads to decreased cardiac glucose uptake and increased dependence on free fatty acid utilization [125]. This metabolic switch impairs oxidative phosphorylation and disrupts mitochondrial function. In the setting of diabetes, impaired mitochondrial oxidative phosphorylation leads to mitochondrial ROS production and mitochondrial dysfunction. Dysfunctional oxidative phosphorylation and increased ROS may impair the anti-inflammatory properties of CRM [126, 127]. Patients with HFpEF have increased circulating and tissue MDMs. MDMs are also metabolically impacted and MDMs are key mediators of HFpEF and diabetic cardiomyopathy, in part due to pro-inflammatory signaling involving NF-κB and the NLRP3 inflammasome [128].

In lean mice, adipocytes transfer their mitochondria to macrophages in the adipose tissue; however, in obesity, increased long-chain fatty acids prevent mitochondria from being captured by macrophages. Thus, in obese mice, the presence of long-chain amino acids results in oxidatively damaged, adipocyte-derived mitochondria being released into the circulation so they can be taken up by distant organs, such as the heart, to exert anti-inflammatory effects [129–131]. It is possible that part of the protective effects of mitochondrial transfer is to preserve macrophage cellular metabolism, as cardiac macrophages can leverage mitochondria from distant adipose tissue and from cardiomyocytes and other cells within the heart [65, 132]. The role of mitochondrial transfer to cardiac macrophage function is a topic of considerable interest. The cardio-metabolic stressors driving HFpEF lead to transcriptomic and epigenetic reprogramming. One form of reprogramming of myeloid cells is termed “trained immunity” whereby myeloid cells may demonstrate amplified inflammatory responses with restimulation such as progression of obesity, hyperglycemia, and diabetes [4, 133–135]. HIF1α, succinate, aldosterone are important mediators of trained immunity for MDMs [136–139]. Additively, the ability of free fatty acids to promote CRM IL-10 production is impaired in cardiometabolic syndromes [122]. Together, each of the above mechanisms (impairment of CRMs and potentiation of MDMs) promote diastolic dysfunction.

Cardio-metabolic Response to Exercise

Many of the risk factors for HFpEF are related to sedentary lifestyles. Regular exercise decreases the incidence of obesity and diabetes and reduces cardiovascular mortality [140]. Recent meta-analyses have shown that exercise training significantly decreases biomarkers of cardiometabolic risk in patients with diabetes, including adiponectin, fasting insulin, TNFα, IL-6, and C-reactive protein [141]. Exercise itself has many beneficial effects including improved insulin sensitivity, preservation of endothelial cell function and protective effects on blood pressure and lipid modulation. Exercise can range from leisurely to “extreme” and the benefits on cardiac health follow a curvilinear or “reverse-J” relationship where benefits are realized up until extreme exercise, at which point coronary artery calcification, myocardial fibrosis, and electrical instability occur [142, 143].

Exercise is categorized by its intensity and duration ranging from low to high intensity and from short to prolonged duration. Each of these categories of exercise imposes distinct metabolic demands and can lead to chronic metabolic remodeling. Fatty acids act as the primary fuel during low-moderate intensity exercise with hormone- activated lipolysis increasing free fatty acids [144, 145]. Triacylglycerol and lactate increase which are important determinants of cell signaling and gene expression [146, 147]. Lactate itself can be used for myocardial metabolism but can cause macrophage death [147, 148]. During exercise, glucose metabolites simulate pro-hypertrophic kinases such as mTOR and AMPK [146]. Fatty acid binding protein 3 and 4 which facilitate glucose and free fatty acid uptake in the heart are increased during exercise [149]. In the setting of regular exercise, decreases in glycolytic rate promote cardiac growth through Cepbp and Cited4 leading to reversible cardiac remodeling or the “athlete’s heart” [146]. Compared to low intensity exercise, high intensity exercise leads to increased malondialdehyde which is the main by-product of lipid peroxidation caused by increased ROS and superoxide dismutase activity. During high intensity exercise, increased phosphatidylcholine levels lead to impairment of fatty acid utilization and accumulation of 1-stearoyl −2 oleoyl-sn-glycero 3 phosphocholine which further accelerates the production of l-carnitine, betaine, lyso-phosphatidylcholine and trimethylamine-N-oxide leading to myocardial damage. During prolonged endurance exercise, glucose catabolism in the heart is decreased by diminishing activity of phosphofructokinase (PFK) [150]. At the same time, skeletal muscle levels of HIF1α are increased [151].

Macrophage Metabolic Responses to Exercise

Routine exercise decreases production of peripheral leukocytes [152, 153] through. inducing hematopoietic stem and progenitor cell (HSPCs) quiescence [153]. Bone marrow of exercising mice express HSPC quiescence genes Cxcl12, Vcam1, Kitl, Angpt [153], potentially driven by reduced adipose-tissue derived leptin signaling [153]. Moderate exercise promotes favorable macrophage behaviors including phenotype switching through upregulation of AMAC1, CD14, MR, IL-4, PGC-1 and downregulation of inflammatory MCP-1, IL-6, TNFα [154]. Moderate exercise promotes macrophage driven angiogenesis and wound healing in mice [155, 156] and activates CRM to secrete IL-10 to promote cardiac protection [157]. Exercising mice that genetically lack IL-10 in CRMs develop impairment in ejection fraction, cardiac hypertrophy, and fibrosis when additively given isoproterenol. In CRMs, arginase 1 is induced and produces urea, polyamines, and ornithine, which are important for wound healing [158]. Exercise has a number of other anti-inflammatory and pro-resolution effects. After only four weeks of voluntary wheel exercise, mice demonstrated increased levels of SPMs (RvEV1, RvD1, MaR1) in macrophages [159]; conversely, ghrelin, which inhibits chronic intermittent hypoxia induced migration of macrophages by suppressing ROCK2, is reduced with exercise [160]. Exercise induces PPARγ activation and phagocytic function [156] and leads to a reduction of β2-adrenergic receptors in macrophages, leading to reduced inflammation possibly through inhibition of NF-κB [161]. Chronic exercise reduces TLR4 signaling and suppresses IL-12 production. After cessation of activity, long-term maintenance of anti-inflammatory phenotype is temporarily preserved through epigenetic alterations [153]. The contribution of high intensity or extreme exercise to macrophage functions are less well understood.

Concluding Remarks

Macrophages are critical mediators of development, health, and disease. Macrophages interface with their local niche within the heart and are impacted by both local and systemic environmental conditions. Metabolic demands change over life from a hypoxic environment in utero to oxygen rich in post-natal life. Pathologic (IRI, diabetes, obesity) and physiologic (exercise) mechanisms are added contributors to macrophage metabolism and function. Dissecting out the relative contributions of CRM and MDM are some of several issues critical to understand as we aim to take a personalized approach to patient care (see Outstanding Questions).

Outstanding Questions.

• What is the unique role of CRMs and MDMs and their respective subpopulations in cardiovascular health and physiology, disease pathogenesis, and tissue repair?

• How do exercise intensity, duration, and type impact cardiovascular metabolism and the immune-stromal landscape?

• Does excessive exercise skew metabolism to prioritize activation of pro-inflammatory macrophages?

Highlights:

• Cardiac macrophages are comprised of two distinct lineages: embryonic-derived macrophages and definitive monocyte-derived macrophages.

• Embryonic-derived cardiac macrophages are fueled through oxidative phosphorylation, suppress inflammation, and are essential regulators of myocardial regeneration, coronary angiogenesis, adaptive tissue remodeling, electrical conduction, and metabolism.

• Monocyte-derived macrophages are recruited to the injured and diseased heart where they differentiate into numerous subsets including inflammatory populations that contribute to adverse remodeling and heart failure.

• Pro-inflammatory subsets of monocyte-derived macrophage favor glycolysis over oxidative phosphorylation.

• Metabolic pathways and metabolites modulate monocyte fate specification and myocardial inflammatory responses. Dysregulation of metabolic pathways, for instance, in diabetic patients, perturbs macrophage homeostasis and contributes to progression of disease.

• Exercise may represent one approach to shape macrophage metabolism and function.

Glossary

Cardiac Resident Macrophages

Macrophages that seeded the heart during embryonic development and self-renew independent of ongoing monocyte recruitment

Heart Failure with Reduced Ejection Fraction

Occurs when the left ventricle of the heart does not contract normally. Generally, a cut-off of a left ventricular ejection fraction below 40% is used. Most approved heart failure therapies are for patients who fall within this category

Heart Failure with Preserved Ejection Fraction

Occurs when the left ventricle of the heart does not relax normally. This reduces the ability of the heart to fill with blood. Generally, a cut-off of a left ventricular ejection fraction above 50% is used. A growing understanding that more people than previously recognized fall within this category. There are limited, but growing, therapies for patients with heart failure with preserved ejection fraction. A definition of heart failure with mid-range ejection fraction is a newer concept where left ventricular ejection fraction is between 41–49%.

Monocyte Derived Macrophages

Macrophages that derived from definitive hematopoietic stem cells and originate from the bone marrow and spleen and enter the heart during times of injury