Obesity, Hypertension, and Cardiac Dysfunction: Novel Roles of Immunometabolism in Macrophage Activation and Inflammation

Abstract

Obesity and hypertension (HTN), which often coexist, are major risk factors for heart failure (HF) and are characterized by chronic, low-grade inflammation, which promotes adverse cardiac remodeling. While macrophages play a key role in cardiac remodeling, dysregulation of macrophage polarization between the pro-inflammatory M1 and anti-inflammatory M2 phenotypes promotes excessive inflammation and cardiac injury. Metabolic shifting between glycolysis and mitochondrial oxidative phosphorylation (OXPHOS) has been implicated in macrophage polarization. M1 macrophages primarily rely on glycolysis while M2 macrophages rely on the tricarboxylic acid (TCA) cycle and OXPHOS; thus, factors that affect macrophage metabolism may disrupt M1/M2 homeostasis and exacerbate inflammation. The mechanisms by which obesity and HTN may synergistically induce macrophage metabolic dysfunction, particularly during cardiac remodeling, are not fully understood. We propose that obesity and HTN induce M1 macrophage polarization via mechanisms that directly target macrophage metabolism, including changes in circulating glucose and fatty acid substrates, lipotoxicity, and tissue hypoxia. We discuss canonical and novel pro-inflammatory roles of macrophages during obesity-HTN induced cardiac injury, including diastolic dysfunction and impaired calcium handling. Finally, we discuss the current status of potential therapies to target macrophage metabolism during HF, including anti-diabetic therapies, anti-inflammatory therapies, and novel immunometabolic agents.

INTRODUCTION

Cardiovascular diseases (CVD) kill almost 1 million US citizens every year.¹ Obesity and hypertension (HTN), which are associated with the metabolic syndrome, are highly prevalent in CVD patients, and contribute to the pathogenesis of heart failure (HF) with preserved ejection fraction (HFpEF) as well as heart failure with reduced ejection fraction (HFrEF), which most commonly manifests after myocardial infarction (MI).^1–3 However, neurohormonal therapies aimed at controlling HTN and excessive sympathetic nervous system (SNS) and renin-angiotensin aldosterone (RAAS) activation have proven more efficacious in treating HFrEF than HFpEF, and new therapies for treatment of HFpEF that reduce morbidity and mortality are needed.⁴

Obesity, especially when associated with excess visceral adiposity, is a major risk factor for HTN and HF.^{5, 6} HTN and obesity as well as multiple metabolic disorders and inflammation are highly prevalent in HFpEF and neurohormonal therapies are largely ineffective in these patients.^{4, 7} Epidemiologic studies suggest that 65–78% of the risk for primary HTN may be attributed to obesity,⁸ and long-standing HTN is an important risk factor for development of chronic kidney disease (CKD) and HF.^{3, 5, 9} Obesity-induced HTN is characterized by excessive RAAS and SNS activation which is due, in part, to changes in circulating adipokines such as leptin.^{8, 10} Both obesity and HTN promote left ventricular (LV) pressure and volume overload, eventually leading to HF.^{3, 8, 11} While animal models typically focus on one type of stress, such as HTN, obesity, or diabetes, these cardiometabolic disturbances are rarely observed in isolation during human HF.^{3, 8} In fact, obese-hypertensive HFpEF is becoming more appreciated as a unique entity, as obesity not only raises blood pressure but also increases LV volume load and exacerbates inflammation.² HTN, in the absence of obesity, primarily causes LV pressure overload, leading to thickened ventricular walls (concentric hypertrophy); however, obesity-induced HTN is initially associated with volume overload, increased cardiac output, LV dilation (eccentric hypertrophy) as well as concentric hypertrophy.^{8, 12} Thus, mechanical stress on the heart is especially increased in obesity-induced HTN.^{2, 5, 8}

Although the heart initially adapts to chronically increased loads by undergoing compensatory hypertrophy and remodeling, decompensation may eventually occur as the heart dilates and loses effective contractile function.^{9, 13} Thus, severe diastolic dysfunction, leading to HFpEF, often precedes systolic dysfunction and HFrEF, although many HFpEF patients never develop HFrEF.^{9, 13} In patients with HF, obesity is associated with worse HF symptoms and greater inflammation.^{2, 14} Importantly, the inflammatory and abnormal metabolic milieu in obese patients is thought to be a key amplifier of cardiac injury during HTN.⁶

The systemic and cardiac inflammation associated with obesity-induced hypertension and adverse cardiac remodeling is largely mediated by monocytes/macrophages, which assume a pro-inflammatory (M1) phenotype during HF^15–22; thus, curbing inflammation and macrophage polarization may be a promising therapeutic option for HF, especially obesity-associated HFpEF.^{20, 23, 24} Clinically, the number of circulating pro-inflammatory monocytes and monocyte/macrophage-derived cytokines correlates with worse outcomes in HF patients.^{25, 26}

This review focuses on the role of macrophages in mediating or amplifying cardiac injury during obesity and HTN which often coexist. We also discuss potential mechanisms of obesity- and HTN-induced changes in macrophage metabolism as major drivers of pro-inflammatory activation, specifically in terms of changes in substrate utilization and tissue hypoxia. Given the plethora of metabolic changes that occur during obesity-HTN, metabolic reprogramming appears to be the critical driver of macrophage polarization in obesity-HTN-associated HF. Although we refer to M1/M2 phenotypes throughout this manuscript, it is important to note that this dichotomy is largely over-simplified and based mainly from in vitro responses to polarizing stimuli.³ In reality, multiple macrophage subtypes exist with overlapping functions and phenotypes, and are exposed to hundreds of different stimuli which are integrated by complex signaling pathways to generate a defined output.²⁷ As our knowledge of macrophage metabolism grows, so too will our understanding of how this important cellular function plays a role in cardiac macrophage heterogeneity.

CARDIAC METABOLIC CHANGES IN OBESITY AND HYPERTENSION

Cardiac metabolic reprogramming, particularly in myocytes, has been heavily implicated in cardiac adaptation to injury.^{12, 18, 28} While the healthy heart relies mainly on mitochondrial oxidative phosphorylation (OXPHOS) metabolism of fatty acids (~70%) for its energy demands, during decompensated HF the heart relies predominantly on glucose metabolism (glycolysis).^{18, 29} Metabolic shifts during compensatory hypertrophy are less clear, and may depend on the type of injury, as well as the animal model.³⁰ In obese and/or diabetic subjects, cardiac fatty acid oxidation (FAO) is favored over glycolysis and glucose oxidation as the main source of energy production, mainly as a result of increased substrate availability (circulating lipids) and impaired ability to use glucose due to insulin resistance.^31–33 Despite increased cardiac lipolysis in obesity, the heart still becomes overloaded with lipids due to excess substrates that exceed energy demands, leading to lipotoxicity, oxidative stress, and mitochondrial dysfunction.^{34, 35} The role of altered cardiac metabolism during obesity HTN in humans is still unclear, and appears to depend on the severity of HTN and cardiac hypertrophy.³⁶ Regardless of which metabolic pathway predominates, the failing heart is characterized by metabolic inefficiency and oxidative stress.³⁷

Another defining feature of HF is mitochondrial dysfunction.^{12, 18} Obesity and HTN synergistically impair cardiac mitochondrial biogenesis and function in swine models.³⁸ Thus, while the failing heart may switch to predominantly glycolysis due to mitochondrial injury, obesity and insulin resistance may worsen cardiac injury by impairing glucose utilization. Whether non-parenchymal cardiac cell populations such as macrophages diverge in their metabolic responses to cardiac injury remains poorly understood. This question is important as therapies targeting myocyte metabolism with unknown effects on other cardiac cells continue to be prescribed.^{4, 39–41}

MECHANISMS OF OBESITY HYPERTENSION-INDUCED INFLAMMATION AND M1 MACROPHAGE POLARIZATION

The inflammatory response is a critical mechanism by which the heart responds to injury to undergo adaptive remodeling.²⁴ However, excessive inflammation can impair this adaptive response and exacerbate cardiac injury.²⁴ Both obesity and HTN lead to chronic systemic inflammation, characterized by excess macrophage activation and M1 polarization, which release pro-inflammatory cytokines that impair cardiac function, and circulating pro-inflammatory cytokines correlate with HF mortality.⁷ Thus, there are overlapping and potentially synergistic mechanisms by which obesity and HTN promote inflammation and M1 polarization. Macrophages “sense” obesity and HTN-induced target organ damage through changes in metabolic substrates, hormones, local inflammatory signaling molecules, mechanical stretch, and hypoxia-induced signaling.

Tissue Hypoxia

One of the most important signaling factors in the metabolic switch from OXPHOS to glycolysis in macrophages is activation of hypoxia-inducible factor-1-alpha (HIF-1α), which promotes transcription of a glycolytic and pro-inflammatory M1 gene profile.^{19, 42} Obesity is associated with enlargement of adipocytes and cell hypoxia which activates HIF-1α in adipose tissue macrophages, stimulating chemokines/cytokines that activate and recruit monocyte reservoirs.⁴³ Deletion of HIF-1α in monocytes/macrophages attenuates adipose tissue inflammation and improves insulin resistance in obese mice.⁴⁴ Obesity may also induce myocardial hypoxia through capillary rarefaction and reduced capillary density, and through myocardial lipid accumulation (fatty heart).^{45, 46} During pressure overload-induced cardiac remodeling (i.e. HTN), increased cardiac work load causes imbalances between cardiac hypertrophy and capillary density that leads to hypoxia.⁴⁷ Thus, obesity and HTN both promote myocardial hypoxia that also affects cardiac macrophages.

HIF-1α can also be stimulated by non-hypoxic mechanisms associated with obesity and HTN, including pro-inflammatory cytokines, hyperglycemia, saturated fatty acid activation of toll-like receptor-4 (TLR4), and oxidized LDL.^48–51 Thus, through effects on cardiac hypertrophy and lipid accumulation, obesity and HTN may synergistically activate macrophage HIF-1α to promote a glycolytic pro-inflammatory phenotype.

Circulating Metabolic Substrates

Obese patients are often diabetic with symptoms such as hyperglycemia and dyslipidemia,^{31, 33} which can alter macrophage phenotype.^{50, 52–54} For example, diabetic patients are prone to bacterial infections and impaired wound healing, which may be due to impairments in pro-reparative/anti-inflammatory M2 macrophage functions.^{52, 55} Macrophages are sensitive to extracellular glucose levels, as hyperglycemia induces an M1 phenotype in macrophages through increased expression of long-chain acyl-CoA synthetase-1, promoting lipid accumulation and lipotoxicity.⁵³ Unlike cardiomyocytes, macrophages are not insulin-sensitive and thus retain the ability to take up glucose during insulin-resistant states such as obesity/diabetes, while myocytes shift to OXPHOS.^{31, 56} Thus, glucose uptake in macrophages appears to be dependent on glucose availability, which feeds directly into glycolysis and promotes M1 polarization.^{54, 56}

Macrophages are also susceptible to protein glycation and formation of advanced glycation end-products (AGEs) due to excess glucose levels, which activates the nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) pathway and production of inflammatory cytokines.^{57, 58} Macrophages from patients with coronary artery disease (CAD), a common co-morbidity of obesity-HTN, display significantly higher levels of glucose utilization and increased interleukin (IL)-1β and tumor necrosis factor-alpha (TNF-α) expression than those from patients with giant cell arteritis (a purely inflammatory vascular disease).⁵⁴ Therefore, during hyperglycemia, macrophages appear to upregulate glucose uptake and utilization and subsequently increase cytokine production.⁵⁴

Changes in circulating lipids can also affect macrophage function and promote lipotoxicity.⁵⁹ Macrophage lipotoxicity has been well characterized in atherosclerosis, often observed during obesity-HTN, in which circulating monocytes become overloaded with lipids such as oxidized low-density lipoprotein (LDL), leading to formation of pro-inflammatory foam cells, and is characterized by endoplasmic reticulum (ER) stress, mitochondrial dysfunction, and apoptosis.⁵⁹ Fatty acids and ceramides, which are both elevated in obesity, also act as signaling molecules and induce M1 polarization by activation of pattern recognition receptors, including TLR4.^{60, 61}

In addition to activating pro-inflammatory signaling pathways, excess free fatty acid uptake in circulating and tissue monocytes/macrophages shifts fatty acid utilization away from OXPHOS towards triglyceride, phospholipid, and ceramide synthesis, promoting lipotoxicity and an M1 phenotype.^62–64 Thus, macrophages are especially sensitive to changes in energetic substrates associated with obesity and the metabolic syndrome, and macrophage substrate metabolism and polarization may be influenced by extracellular availability of substrates, especially increased levels of glucose and lipids.

Lipid Mediators

Fatty acids play a key role in macrophage polarization via formation of small lipid mediators, which can be pro-inflammatory or anti-inflammatory/pro-resolving mediators.^{65, 66} Lipid mediators can act in an endocrine, paracrine, autocrine, and even intracrine manner, mainly through activation of G-protein coupled receptors (GPCRs) in macrophages.⁶⁷ The composition of fatty acids in the diet can have a drastic effect on these lipid mediators.^{67, 68} For example, dietary omega-6 fatty acids such as linoleic acid give rise to inflammatory mediators (e.g. prostaglandins, leukotrienes), while omega-3 fatty acids (e.g. eicosapentanoic acid, docosahexaenoic acid) give rise to resolvins, which promote resolution of inflammation.⁶⁹ Some forms of HTN also increase circulating isoprostane levels, which promote pro-inflammatory monocyte polarization.⁷⁰ Production of these mediators is regulated by macrophage expression of different cyclooxygenase (COX) and lipoxygenase (LOX) enzymes.⁷¹

Pro-inflammatory lipid mediators support NF-κB signaling and cytokine production, and impair anti-inflammatory/anti-oxidant pathways.⁷² A high-fat diet enriched with sunflower oil to mimic obesogenic Western diets dysregulates lipid mediators to favor pro-inflammatory mediators.⁷³ Thus, excess intake of omega-6 and decreased omega-3 fatty acids may also promote M1 polarization during obesity.⁷²

Less is known about the role of lipid mediators in regulating macrophage function during HTN. Different forms of HTN appear to be associated with increased circulating levels of inflammatory lipid mediators, as pulmonary HTN patients have elevated levels of pro-inflammatory prostaglandin-D2 in association with macrophage activation, while anti-inflammatory resolvin-D1 is decreased in patients with carotid atherosclerosis ^{74, 75} In the kidney, macrophages increase expression of cyclooxygenase-2 (COX-2) and prostaglandin-E1 synthase in response to a high-salt diet, and mice lacking COX-2 specifically in hematopoietic cells (which give rise to leukocytes) develop salt-sensitive HTN.⁷⁶ Damaged endothelial cells also synthesize lipid mediators, which promote expression of adhesion molecules and monocyte extravasation into tissues.⁶⁸ Proliferative vascular smooth muscle cells produce pro-inflammatory lipid mediators during vascular hyperplasia, leading to vascular inflammation.⁶⁸ Thus, obesity and HTN can dysregulate the balance between pro- and anti-inflammatory lipid mediators to promote M1 polarization.

Adipokines/Hormones

Adipokines, which are circulating cytokines/hormones released from adipose tissue, also influence macrophage polarization.^77–79 Leptin, an adipokine known for its anorexic effects, also regulates autonomic, cardiovascular, and metabolic functions, and is highly elevated in obese patients,⁸⁰ and contributes to obesity-HTN via SNS activation.⁸ Leptin is a potent activator of innate and adaptive immune cells, and promotes pro-inflammatory cytokine secretion and an M1 macrophage phenotype during obesity.^{77, 78} In contrast, adiponectin is another adipokine that exerts anti-inflammatory actions and promotes an M2 phenotype; however, adiponectin levels are decreased in obesity.^{60, 79} Thus, increased leptin and reduced adiponectin may contribute to M1 macrophage polarization during obesity-HTN.

Brown adipose tissue, which has favorable thermogenic, metabolic, and cardiovascular effects, can also secrete adipokines that affect macrophage polarization.⁸¹ CXCL14 released from brown adipocytes promotes an M2 macrophage phenotype, and protects obese mice from glucose intolerance and insulin resistance.⁸¹

The RAAS, which plays a central role in the development of obesity-HTN, also promotes M1 polarization and release of inflammatory monocytes from the spleen via angiotensin II (Ang II) activation of the angiotensin-type I receptor.^{8, 82, 83} Circulating monocytes polarize to an M1 phenotype in response to circulating catecholamines and aldosterone.⁸⁴

Nervous System Actions on Leukocyte Reservoirs

During cardiac injury, macrophages are derived largely from infiltrating monocytes from circulating and splenic reservoirs.^85–90 Obesity-induced HTN is often characterized by chronic SNS activation.⁸ The SNS can promote M1 polarization via direct innervation of the spleen, which has become recognized as the neuroimmune axis.^{21, 91} Through adrenergic pathways, sympathetic nerves stimulate activation of quiescent splenic macrophage reservoirs and their polarization towards an M1 phenotype, and can also promote extramedullary hematopoiesis, or production of new monocytes.^{21, 85} The neuroimmune axis may also contribute to HTN through recruitment of immune cells that cause further vascular and renal injury.⁹²

Conversely, the parasympathetic nervous system promotes M2 polarization, as vagal stimulation of the spleen increases M2 macrophage numbers via acetylcholine neurotransmission through the α7-nicotinic acetylcholine receptor.^{21, 93} Since obesity-HTN is associated with reduced parasympathetic activity, especially in the heart, ^{94, 95} this may shift macrophage polarization toward M1.Furthermore, non-neuronal cholinergic signaling plays an important role in macrophage polarization, as acetylcholine released locally from cardiomyocytes, endothelial cells and T cells also activate muscarinic acetylcholine receptors on macrophages to promote M2 polarization.^{13, 94, 96}

SOURCES OF CARDIAC MACROPHAGES DURING OBESITY-HTN

Resident Cardiac Macrophages

Resident macrophages exist in the healthy heart in low numbers, where they typically assume an M2 phenotype.^{85, 87, 97} While resident cardiac macrophages have often been considered protective, recent studies have demonstrated these embryonic-derived M2-like macrophages are lost with aging and replaced with CCR2+ monocyte-derived macrophages, even in the absence of injury.⁹⁷ In the pressure-overloaded mouse heart, resident macrophages initially promote myocardial adaptive remodeling to mechanical stress, while infiltrating CCR2+ monocytes promote maladaptive remodeling during the transition to decompensated HF.⁹⁸

Epicardial Adipose Tissue Macrophages

Macrophages are by far the most abundant immune cell in adipose tissue.⁹⁹ Epicardial adipose tissue (EAT) is the layer of adipose tissue surrounding the heart, and shares the same myocardial microcirculation.^100–102 In the healthy heart, EAT plays a protective role, as it can provide the heart with fatty acids for energy, or conversely serve as a buffer for lipid overload.¹⁰³ Like the spleen, EAT is a rich source of macrophages, neutrophils, and lymphocytes, which may protect the healthy heart from infection.¹⁰¹ M2 macrophages support the protective effects of EAT, as myokines released from the heart muscle during exercise act directly on EAT macrophages to promote an M2 phenotype.¹⁰³ However, during cardiac injury, EAT can be a major source of M1 macrophages. As previously discussed, adipose tissue hypoxia during obesity is a major factor in M1 macrophage polarization. This hypoxia also extends to EAT, leading to activation of M1 macrophages that can easily infiltrate the myocardium.^{46, 104, 105} In mice, EAT is a major contributor to inflammation during cardiac injury after MI,¹⁰¹ and in MI patients, increased EAT thickness is associated with visceral obesity and cardiac fibrosis.¹⁰⁰ EAT dysfunction is also observed in HFpEF, as obese HFpEF patients have thicker EAT which may prevent the heart from appropriately filling during obesity-induced volume overload.^{2, 14} Thus, EAT is another leukocyte reservoir for the injured heart, and may aggravate injury in obese-HTN patients.

Splenic and Bone Marrow-Derived Macrophages

The splenic-cardiac axis has been implicated during injury, particularly in MI^{71, 90} but more recently in HTN and hypertensive cardiac injury.^{22, 91, 98} After MI, splenic monocyte recruitment depends on Ang II signaling,⁸³ which may be elevated during obesity-HTN.⁸ While obesity leads to splenomegaly and splenic macrophage expansion,¹⁰⁶ its role in the splenic-cardiac axis remains to be determined.

Like the spleen, bone marrow is a major site of hematopoiesis and production of monocytes that can infiltrate the myocardium during injury.⁸⁵ MI induces proliferation of bone marrow hematopoietic stem cells (HSCs), which rely on adrenergic signaling to enter the circulation.¹⁰⁷ Bone marrow-derived HSCs can also enter the spleen where they contribute to splenic hematopoiesis.¹⁰⁸

Salt Sensing

While plasma sodium (Na+) concentrations are tightly controlled by thirst and antidiuretic hormone-mediated renal mechanisms, Na+ can accumulate in tissues, especially during salt-sensitive HTN.^{109, 110} In antigen-presenting cells (i.e. macrophages), exposure to high concentrations of Na+ promotes expression of M1 markers, including IL-1β, via the salt sensing serum glucocorticoid kinase (SGK1).¹¹⁰ This salt sensing ability of macrophages seems to have a conserved role in wound healing, as salt accumulates in wounds and thus promotes macrophage activation and healing.¹⁰⁹ Thus, in addition to activation of the SNS and RAAS, HTN may lead to M1 polarization by promoting a salty microenvironment, especially in salt-sensitive forms of HTN which are often associated with obesity, diabetes and impaired kidney function.¹¹¹

Cardiomyocyte Injury and Death

Injured myocytes also promote localized cardiac inflammation. During severe cardiac injury (e.g. MI), necrotic myocytes release damage-associated molecular patterns (DAMPs) that recruit and activate macrophage-mediated inflammatory responses.^{85, 86} Mechanical overload (i.e. during HTN) can also stimulate myocytes to release pro-inflammatory cytokines.^{112, 113, 114} Damaged mitochondria in cardiomyocytes, a hallmark of HF, can also release intracellular DAMPs such as mitochondrial DNA that promote myocyte release of pro-inflammatory chemokines and cytokines.¹¹⁵ Thus, direct cardiac injury during obesity and HTN promotes local inflammation within the cardiac microenvironment.

Endothelial Damage

Both obesity and HTN promote vascular/endothelial injury, through overlapping and unique mechanisms, leading to release of inflammatory signals from the injured vasculature and subsequent monocyte recruitment and M1 polarization.^{21, 32} Excess perivascular fat in obesity secretes pro-inflammatory cytokines that promote endothelial inflammation and injury.¹¹⁶ During HTN, pressure-induced shear stress and mechanical stretch promotes endothelial injury and expression of cytokines, chemokines, and adhesion molecules that enhance leukocyte adhesion.^{117, 118} HTN-induced endothelial injury also increases reactive oxygen species (ROS) production and impairment of nitric oxide (NO) signaling, which may attract nearby macrophages.¹¹⁷ Thus, coronary microvascular endothelial dysfunction and injury may be an important risk factor in HF pathophysiology, particularly HFpEF.^{16, 38}

Mechanical Stretch

As mentioned, mechanical stretch due to chronic HTN plays a critical role in mediating cardiac and vascular injury. Macrophages can also sense mechanical stretch through interactions with the ECM.¹¹⁹ Cyclical mechanical stretch in alveolar macrophages, for example, promotes inflammasome activation and IL-1β secretion, which contributes to lung injury.¹²⁰

Progression of Obesity-HTN Induced Macrophage Activation

We have discussed numerous mechanisms by which obesity-HTN gives rise to pro-inflammatory M1 macrophages in the body and heart; however, these events likely do not occur simultaneously but rather sequentially. For example, obesity-HTN occurs frequently as young as early childhood¹²¹, yet severe complications (i.e. HF) may not develop until later in life.^{8, 122} Thus, it is likely that during metabolic syndrome, macrophage activation and inflammation are initiated by diet composition and obesity, eventually leading to hyperglycemia and insulin resistance, SNS and RAAS activation and HTN, endothelial damage, and HF; along this spectrum, excessive M1 macrophage activation exacerbates each of these pathologies to create a potential vicious cycle (Figure 1).

Figure 1. Mechanisms of pro-inflammatory M1 macrophage polarization over the obesity-HTN disease spectrum.

Obesity and its associated complications promote inflammation and macrophage activation, which in turn exacerbates later complications of obesity including SNS/RAAS activation, vascular injury, and heart failure.

ROLES OF MACROPHAGES IN MEDIATING OBESITY HYPERTENSION-INDUCED CARDIAC INJURY AND DYSFUNCTION

Macrophages play a critical role in LV inflammation and remodeling by releasing pro-inflammatory cytokines and growth factors that act on neighboring resident cells, and phagocytosing necrotic and/or apoptotic myocytes (Figure 2). Resident macrophages exist in the healthy heart as a heterogenous pool of distinct subsets, and promote more efficient healing and less damage than infiltrating monocyte-derived macrophages.^{88, 97, 107} In the adult heart, however, resident macrophages lose their ability to self-renew, and after injury are largely replaced by circulating and splenic monocytes, which promote less efficient healing.^{85, 123} M1 macrophages promote adverse cardiac remodeling through effects on myocyte function, fibrosis, and angiogenesis.

Figure 2. Mechanisms of macrophage-mediated cardiac remodeling.

During cardiac injury and remodeling, macrophages secrete pro-inflammatory cytokines that impair myocyte systolic and diastolic function, activate nearby fibroblasts to promote fibrosis, and inhibit growth of neighboring endothelial cells to impair angiogenesis and neovascularization.

Role of Macrophages in Myocyte Hypertrophy and Dysfunction

The heart can hypertrophy under different types of stressors. Physiological hypertrophy is induced by stimuli such as exercise or pregnancy and is mediated by Akt-mTOR signaling pathways.¹¹ Pathological hypertrophy, also referred to as compensatory hypertrophy, is characterized by an increase in LV mass without improvement of ventricular function. Pathological hypertrophy is mediated by MAP kinase pathways, involves isoform switching of several sarcomere proteins, as well as increased secretion of natriuretic peptides that initially help relieve cardiac workload.¹¹ While this hypertrophy initially allows the heart to adapt to mechanical stress of pressure and/or volume overload during HTN, the heart can eventually decompensate, and HFrEF develops.¹¹

The major stimulus for pathological hypertrophy is mechanical, as pressure/volume overload is sensed by stretch receptors that activate mechanotransducing signaling pathways cause pathological hypertrophy, while neurohormonal mechanisms (SNS and RAAS activation) play a secondary role, as Ang II-induced cardiac hypertrophy is mediated by renal mechanisms that ultimately lead to increased blood pressure.^{124, 125} Macrophages contribute to pathological hypertrophy through secretion of pro-inflammatory cytokines (TNF-α, IL-1β, and IL-6), which activate MAP kinase and NF-κB signaling pathways and inhibit the Akt-mTOR pathway in neighboring cardiomyocytes.^{11, 126, 127}

Cardiomyocyte calcium handling, a critical component of ventricular function, is typically impaired during injury and HF.¹²⁸ Calcium release from the sarcoplasmic reticulum (SR) via the ryanodine receptor (RyR) and calcium reuptake via SR Ca²⁺ ATPase (SERCA2a) is necessary for appropriate relaxation and is necessary for normal excitation-contraction coupling.^{128, 129} Appropriate calcium handling is also important for mitochondria, in which calcium uptake supports OXPHOS and ATP production and limits excessive ROS production during ventricular contraction.¹³⁰ Macrophages influence expression of calcium-handling proteins in vitro, as M1 macrophage-derived pro-inflammatory cytokines downregulate SERCA2a in mouse embryonic stem cell-derived cardiomyocytes, while M2 macrophages increase expression of SERCA2a, which may help preserve cardiac function following injury.¹²⁶ Macrophage-derived IL-1β directly impairs myocyte calcium currents in diabetic mice.¹³¹ Macrophages also directly couple to cardiomyocytes via gap junctions and mediate electrical conduction in the atrioventricular node in the healthy heart; however, pro-inflammatory macrophages may contribute to arrhythmias during HF via production of IL-1β.¹²⁷

The cardiomyocyte nitric oxide (NO)-protein Kinase G (PKG)-titin axis plays a central role in the development of HFpEF.^{7, 132} Macrophages may influence cardiac remodeling via alterations in NO signaling, which affects compliance of the giant myofibrillar protein titin.^{7, 132} In an obese mouse model of HFpEF, deletion of inducible NO synthase (iNOS), which is mainly expressed by inflammatory cells and upregulated during inflammation, prevented cardiac nitrosative stress and cardiac injury.¹³³ Furthermore, pro-inflammatory cytokines impair production of NO by cardiac microvascular endothelial cells while inducing production of reactive oxygen species such as peroxynitrite.⁷

The major effect of impaired NO signaling in the myocardium is decreased activation of soluble guanylate cyclase and cyclic GMP (cGMP) activation of protein kinase, leading to titin hypo-phosphorylation and increased titin stiffness.⁷ For this reason, targeting the phosphodiesterases (PDE5a and PDE9a) which degrade cGMP is a potentially attractive therapeutic for HFpEF.¹³⁴

Extracellular Matrix/Fibrosis

Cardiac fibrosis, characterized by excessive extracellular matrix (ECM) deposition, contributes to LV stiffening and diastolic dysfunction.¹³⁵ Fibrosis is caused by excess activation of fibroblasts, the primary cell type responsible for ECM synthesis.¹³⁵ HTN has been well characterized for its strong pro-fibrotic effects.^{114, 136} Obesity also promotes cardiac fibrosis, as both animal models of obesity and obese humans exhibit cardiac fibrosis and circulating markers of collagen synthesis, although in many cases it has been difficult to rule out the impact of HTN which often coexists with obesity.¹²

The mechanisms of obesity- and HTN-induced cardiac fibrosis, much like pathological hypertrophy, are primarily due to mechanical stress-induced signaling pathways in cardiac fibroblasts, but macrophage-induced inflammation is also a contributor.¹² Macrophages play key roles in cardiac fibrosis: indirectly by activating fibroblasts, by secreting transforming growth factor-beta 1 (TGF-β1) and IL-10, and by directly secreting ECM proteins.⁸⁶ In the injured heart, a small percentage of infiltrating monocytes also give rise to fibroblast-like cells which secrete collagen, termed myeloid-derived fibroblasts.¹³⁷

Macrophages are a major source of matrix metalloproteinases (MMPs), whose main function is to degrade components of the ECM during cardiac remodeling, which can then be replaced by new ECM components.¹³⁸ After MI, macrophage MMPs quickly degrade infarcted tissue and release necrotic myocytes from the ECM, which are then phagocytosed and replaced with collagenous scar tissue.¹³⁸ Although M1 macrophages have been heavily implicated in cardiac injury, macrophage-derived IL-10, an M2-marker, is a major contributor to fibrosis and diastolic dysfunction in a mouse model of HFpEF, although this may depend on earlier M1 infiltration and proliferation.²² While this response is beneficial in the infarcted LV, which requires a substantial amount of collagen for scar formation to prevent rupture, it is detrimental for the non-infarcted LV.⁸⁵ Thus, excessive monocyte infiltration and M1 polarization may ultimately give rise to excessive M2 activation, which contributes to adverse LV remodeling and fibrosis.

Angiogenesis

Angiogenesis, which broadly describes formation of new vessels, especially capillaries, is a critical adaptation for increasing oxygen and nutrient delivery in the hypertrophied heart.¹³⁹ During HTN, however, sustained increases in pressure cause vascular injury and impair angiogenic capacity, leading to imbalances in myocardial oxygen demand and delivery.¹³⁹ Obesity may also reduce capillary density in the heart, independent of blood pressure effects.¹⁴⁰

Macrophages can promote angiogenesis through secretion of paracrine factors such as vascular endothelial growth factor (VEGF), which stimulates formation of new capillaries via endothelial cell recruitment/proliferation.^{86, 141} VEGF secretion from macrophages is stimulated by HIF-1α as a response that helps restore/enhance blood flow to ischemic tissue.¹⁴² Pro-inflammatory M1 macrophage subsets derived from Ly6C^hi monocytes impair angiogenesis during injury by releasing anti-angiogenic factors such as thrombospondin-1, while resident M2 anti-inflammatory subsets promote angiogenesis via VEGF secretion.⁹⁸ Furthermore, blocking macrophage-derived VEGF promotes cardiomyocyte hypoxia and apoptosis and worsens cardiac function during ischemia.⁹⁸

Indirect Effects of Macrophages

While M1 macrophages can directly promote adverse cardiac remodeling during disease states, they can also exacerbate systemic complications of metabolic syndrome, including insulin resistance and HTN.^{32, 33} Obese mice lacking myeloid-specific HIF-1α display higher inflammation and systemic insulin resistance compared to wild-type mice.⁴⁴ Immune cells, including monocytes/macrophages and lymphocytes, appear to play a highly supportive role in HTN.^{143, 144} Mice lacking monocytes/macrophages or B and T lymphocytes, which co-regulate macrophage function, fail to develop Ang II-induced HTN.^{145, 146} Macrophages may promote HTN indirectly by causing kidney injury,¹⁴³ which exacerbates increased blood pressure via neurohormonal mechanisms during obesity-HTN.^{144, 147}

Role of Macrophages and Inflammation in the Transition to Decompensated HF

The role of macrophages and inflammation in the transition from compensated hypertrophy to decompensated HF is still poorly understood. Pro-inflammatory circulating monocytes and cytokines are elevated in patients with HTN, and continue to rise during symptomatic HF and correlate with worse HF outcomes (Figure 3).^{22, 82, 148} However, in patients with HTN and in hypertensive mouse models, significant myocardial monocyte infiltration and macrophage accumulation does not occur until HF symptoms have developed, indicating that cardiac macrophage expansion is an important feature of the transition to decompensated HF, during which local cytokine production contributes to cardiac injury in addition to circulating cytokines.^{22, 98} Whether this transition is triggered by inflammation, or whether inflammation is simply a consequence of the transition to decompensation, is still unknown. This may be due to the fact that, to date, most anti-inflammatory therapies have been administered in later stages of HF and are aimed at reversing rather than preventing inflammation;^{26, 149} however, targeting IL-1β after MI has been shown to delay rehospitalization due to HF.¹⁵⁰ Thus, additional animal and clinical studies are needed to assess whether anti-inflammatory therapies can prevent or delay the transition to decompensated HF.

Figure 3. Role of macrophages and inflammation during the transition to decompensated HF.

In healthy individuals, the heart is populated sparsely with quiescent resident macrophages (blue). During compensatory hypertrophy, there are increased circulating pro-inflammatory monocytes and cytokines, but no changes in cardiac macrophage numbers. When decompensation occurs, circulating levels of pro-inflammatory monocytes and cytokines are further increased, and circulating monocytes begin infiltrating the injured heart.

POTENTIAL ROLE OF METABOLISM IN MEDIATING MACROPHAGE ACTIVATION DURING OBESITY HYPERTENSION

Role of Metabolism in Supporting Macrophage Polarization and Function

Macrophage polarization is closely tied to changes in glycolytic and OXPHOS metabolism. The canonical immunometabolic paradigm is that pro-inflammatory M1 macrophages rely mainly on glycolysis, while pro-reparative M2 macrophages rely on fatty acid-fueled OXPHOS.¹⁵¹ Overexpression of the primary macrophage glucose transporter, GLUT1, increases glycolysis and decreases mitochondrial OXPHOS, and increases pro-inflammatory cytokine production;¹⁵² likewise, macrophages lacking GLUT1 show increased M2 polarization.¹⁵³ Knockdown of pyruvate dehydrogenase kinase 1 (PDK1), which phosphorylates pyruvate dehydrogenase to inhibit its role in pyruvate-derived acetyl CoA formation, also decreases glycolysis and promotes an M2 phenotype.¹⁵⁴

Macrophage metabolic shifts are not only required for energy demands but also for regulation of pro- and anti-inflammatory signaling pathways.^{155, 156} As discussed later, pro- or anti-inflammatory signals also drive metabolic shifts in a positive feedback mechanism. Although glycolysis produces much less ATP than OXPHOS, it can rapidly be induced and generates ATP more quickly, which is important for immediate macrophage activation and responses during infection or wound healing.¹⁵⁵ Furthermore, the reliance of M1 macrophages on glycolysis promotes several key pro-inflammatory macrophage functions. The first and most obvious is a shift to anaerobic metabolism to adapt to hypoxic environments.¹⁵⁷ Upregulation of glycolysis also increases activation of the pentose phosphate pathway (PPP), which increases NADPH levels for NADPH-oxidase production of antimicrobial ROS (hydrogen peroxide and superoxide) and lipid synthesis, which support pro-inflammatory gene expression.¹⁵⁸ NADPH is also used for biosynthesis of antioxidants such as glutathione to buffer excessive ROS production during inflammation.¹⁵⁵ The glycolytic shift in M1 macrophages is accompanied by a truncated tricarboxylic acid (TCA) cycle within the mitochondria, in which accumulation of succinate promotes HIF-1α stabilization and IL-1β production.¹⁵⁷ Furthermore, glycolytic metabolism can also localize to subcellular structures, such as filopodia and lammelipodia, where it fuels cytoskeletal remodeling and thus allows macrophage migration to injury sites.¹⁵⁹ Glycolytic enzymes can also play important signaling roles.¹⁵⁵ For example, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) is a component of the gamma-interferon-activated inhibitor of translation (GAIT) complex, which can inhibit the translation of several inflammatory mRNAs, while pyruvate kinase M2 also functions as a nuclear HIF-1α co-activator.¹⁵⁵

In contrast, activation of mitochondrial OXPHOS, which is mainly fueled by β-oxidation of fatty acids, regulates anti-inflammatory M2 macrophage functions.¹⁵⁵ Shutting down glycolysis increases pyruvate entry into the mitochondria and re-establishes the TCA cycle, decreasing succinate accumulation and thus decreasing HIF-1α activation.¹⁵⁷ Re-establishment of the TCA cycle increases NADH and FADH₂ entry into the ETC, increasing OXPHOS.¹⁵⁷ Furthermore, an intact TCA cycle increases production of alpha-ketoglutarate, which serves as a co-factor for epigenetic activation of M2 genes.¹⁶⁰ M2 macrophages also exhibit increased expression of uncoupling protein 2, which uncouples OXPHOS and mitigates excessive mitochondrial ROS production, which promote expression of M1 genes.¹⁵⁵

While the glycolysis-OXPHOS paradigm has been used to define M1 and M2 macrophage polarization, it is somewhat oversimplified, much like the M1/M2 paradigm. For example, glycolytic M1 macrophages also require electron transport chain (ETC) complex I activity, albeit to generate ROS rather than ATP.¹⁵⁶ In certain situations, M2 macrophages may continue to use glycolysis as well as OXPHOS.¹⁵⁸ Furthermore, different pro-inflammatory stimuli mediate different metabolic responses,¹⁵⁸ and as noted before, different disease states are characterized by different immunometabolic profiles.⁵⁴ In obese mice, adipose tissue macrophages assume an M1 phenotype and show both increased glycolytic and OXPHOS metabolism, while peritoneal macrophages do not alter their metabolism, suggesting that the microenvironment drives immunometabolic adaptations during obesity.¹⁹ Furthermore, while obesity increases mitochondrial fatty acid uptake and OXPHOS in M1 macrophages, it also reprograms the ETC towards ROS (superoxide) rather than ATP production, which supports pro-inflammatory gene expression.¹⁶¹

Macrophage Metabolic Signaling Axes

Because macrophage metabolism is so tightly coupled to pro and anti-inflammatory processes, it is highly regulated by several intracellular signaling pathways (Figure 4). Many of the mechanisms that directly affect macrophage metabolism are relevant to obesity-HTN, including elevated glucose/fatty acid levels and tissue hypoxia. HIF-1α is a master regulator of glycolytic metabolism in macrophages.^{86, 151, 152} Once activated by oxygen-sensitive prolyl hydroxylase or pro-inflammatory stimuli and stabilized by succinate, HIF-1α promotes transcription of several glycolytic genes, including GLUT1 (Slc2a1) and PDK1, as well as pro-inflammatory genes.^{49, 50, 151, 159} M1 macrophages can also shut off oxidative metabolism during inflammation.¹⁶¹ Activation of CD36 by extracellular lipid ligands, such as oxidized LDL, stimulates NF-κB to downregulate ETC components and promotes mitochondrial ROS production.¹⁶¹

Figure 4. Intracellular mechanisms of M1 macrophage metabolic reprogramming during metabolic syndrome.

During obesity-HTN, macrophages sense increases in extracellular glucose, pro-inflammatory fatty acids, hypoxia, and neurohormonal factors, leading to activation of pro-inflammatory NF-κB and HIF-1α pathways, which promote glycolytic and pro-inflammatory gene expression, leading to an M1 phenotype. Glycolysis supports activation of the pentose phosphate pathway (PPP), which generates NADPH to support pro-inflammatory prostaglandin/leukotriene synthesis. Pyruvate entry into the mitochondria is also reduced, leading to a truncated tricarboxylic acid cycle that leads to accumulation of succinate, which supports HIF-1α activation of pro-inflammatory/glycolytic gene expression. Conversely, the anti-inflammatory PPARγ-PGC1β axis, which promotes OXPHOS, is decreased during obesity due to decreases in anti-inflammatory fatty acids such as unsaturated/omega-3 fatty acids. Blue arrows indicate activation, orange arrows indicate inhibition. ACh—acetylcholine, Ang II—angiotensin II, ETC—electron transport chain, FAO—fatty acid oxidation, FAS—fatty acid synthesis, GLUT1—glucose transporter 1, HIF-1α—hypoxia inducible factor-1 alpha, LTNs—leukotrienes, NADPH—nicotinamide adenine dinucleotide phosphate, NE—norepinephrine, NF-κB—nuclear factor kappa-light-chain-enhancer of activated B cells, PGs—prostaglandins, PDH—pyruvate dehydrogenase, PDK1—pyruvate dehydrogenase kinase 1, PGC1β—PPAR gamma coactivator 1 beta, PPARγ—peroxisome proliferator activated receptor gamma, PPP—pentose phosphate pathway, ROS—reactive oxygen species, TCA—tricarboxylic acid cycle.

Peroxisome proliferator activated receptor-gamma (PPAR-γ) is a nuclear receptor which is activated by fatty acids and other lipid ligands, and promotes expression of genes for fatty acid (β) oxidation.¹⁶² While adipocytes are the major cell type expressing PPAR-γ, macrophages also highly express PPAR-γ which regulates fatty acid metabolism and expression of anti-inflammatory/M2 genes.^{163, 164} PPAR-γ stimulates cholesterol efflux and decreases lipotoxicity in macrophages.¹⁶⁴ PPAR-γ is typically activated by unsaturated fatty acids and omega-3 fatty acid-derived eicosanoids, which are decreased during obesity.¹⁶² Macrophage PPAR-γ is also required for HTN development in some experimental models.¹⁶⁵ PPAR-γ mediates M2 polarization by interleukin-4 (IL-4), a potent anti-inflammatory cytokine.¹⁶³ This action is mediated by STAT6, which promotes PPAR-γ activation via expression of the PPAR-γ coactivator-1β (PGC-1β), an inducer of mitochondrial biogenesis.^{99, 166} Thus, the IL-4/STAT6/PPAR-γ axis is critical in promoting the shift towards oxidative metabolism and an M2 phenotype.

Role of Macrophage Metabolism During Cardiac Remodeling

While the importance of macrophages in cardiac remodeling has been extensively studied,^{20, 25, 87, 88, 97} the role of macrophage metabolism has not been thoroughly investigated. In mice, glycolytic genes such as GAPDH are strongly upregulated in macrophages found in the infarcted region 1 day after MI, while mitochondrial genes such as succinate dehydrogenase are increased 3 days after MI.⁸⁶ These changes reflect the microenvironment, which is hypoxic at day 1, but becomes re-oxygenated at day 3 due to vasculogenesis.^{86, 141} In a rat model of MI, inhibiting glycolysis by 2-deoxyglucose (2-DG) administration decreased cardiac macrophage glycolysis and inflammation, and improved LV function.¹⁶⁷ After MI, macrophage efferocytosis (phagocytosis of necrotic myocytes) increases intracellular fatty acid supply and thus fuels mitochondrial fatty acid oxidation and an M2 phenotype.¹⁶⁸ Circulating monocytes from CAD patients display higher levels of glycolytic metabolism, and maintain this phenotype after differentiation into tissue macrophages.⁵¹ Thus, it remains critical to understand how monocytes/macrophages that are already primed toward glycolytic metabolism respond to pro-M2/OXPHOS signals in the injured heart in obese-HTN patients.

In this review, we focused on the M1/M2 paradigm for the sake of simplicity; however, it is important to note that this classification is largely oversimplified.¹⁶⁹ In the case of MI, these extreme phenotypes reflects the extreme nature of ischemic injury to the heart, which undergoes complete hypoxia followed by scar formation. However, while the “M2” phenotype may be beneficial for scar formation in the ischemic heart, this may not be true for non-ischemic cardiac injury. In transverse aortic constriction-induced pressure overload in mice, resident macrophage proliferation and accumulation occurs during adaptive hypertrophy, while pro-inflammatory monocyte infiltration only occurs during the transition to decompensated HF.^{98, 114} In mice with HTN and diastolic dysfunction, M2 polarization contributes to cardiac fibrosis, although this depends initially on M1 infiltration and expansion.²² Importantly, multiple phenotypes along the M1/M2 spectrum exist, each with potentially different functions, phenotypes, and markers.^{85, 169} In the mouse heart, resident macrophages, which are negative for C-C chemokine receptor 2 (CCR2), can be further subdivided based on major histocompatibility complex II (MHC II) expression.⁸⁹ Furthermore, cardiac monocyte-derived macrophages can also be derived from infiltrating Ly6C^hi (i.e. M1-like) or Ly6C^lo (M2-like) monocytes, which express CCR2.^{98, 170} Thus, metabolic profiles may differ not just based on the M1 and M2 paradigm, but also resident macrophages versus infiltrating monocytes and their subsets. For now, the immunometabolism field is still young and largely based on M1/M2 phenotypes; as the field grows, so will the understanding of metabolic processes of the numerous cardiac macrophage subsets.

Potential Therapies to Target Macrophage Metabolism During HF

There currently exists no therapeutic strategy to directly target macrophage metabolism during disease states. Cardiomyocyte metabolism has been an area of interest for treating HF, but has not translated successfully to the clinic.⁴¹ Trimetazidine (TMZ), which is used to treat angina in HF patients, directly modulates myocardial metabolism by inhibiting fatty acid oxidation.³⁹ Although TMZ also reduces inflammatory markers in HF patients, whether this is a direct effect is unknown as TMZ is a fatty acid oxidation inhibitor and would thus be expected to impair M2 polarization.^{40, 171} Other drugs targeting metabolism have also been used with success to reduce CVD events, albeit through indirect mechanisms, and may also have anti-inflammatory/immunometabolic effects. Thiazolidinediones (TZDs), a group of drugs used for treating diabetes mellitus in patients with CVD, have pronounced anti-inflammatory effects through activation of PPAR-γ, which promotes M2 polarization.^{4, 59}

Anti-inflammatory therapies have recently gained more support for treating HF.²⁵ Anti-TNFα therapies have produced mixed results in HF patients,¹⁴⁹ but anti-IL-1β therapy has shown clinical success in HFrEF patients.¹⁵⁰ However, IL-1β inhibition was not efficacious in HFpEF, possibly due to the fact that IL-1β mediates acute inflammation, and HFpEF is a disease of chronic inflammation.⁴ Other drugs used in patients with obesity-HTN, diabetes and HF may partially work via immunometabolic mechanisms. For example, in addition to its anti-hyperglycemic and cardioprotective actions, metformin also inhibits NF-κB and IL-1β secretion, potentially through inhibition of mitochondrial complex I.^{4, 156}

Sodium-glucose cotransporter-2 (SGLT2) inhibitors have been shown to reduce HF events, especially HFrEF, but the mechanisms remain elusive.¹⁷² SGLT2 inhibition inhibits macrophage NLRP3 inflammasome activity and thus prevent inflammatory cytokine release,¹⁷³ and alleviates renal and hepatic inflammation in obese and diabetic mice.¹⁷⁴ While the anti-inflammatory mechanisms of SGLT2 inhibitors are still largely unknown, they have been linked to decreased plasma glucose and increased plasma beta-hydroxybutryate, a ketone body which promotes an M2 phenotype.¹⁷⁵ SGLT2 inhibition with empagliflozin directly impairs pro-inflammatory cytokine release from cultured macrophages through an unknown mechanism.¹⁷⁶ Thus, the anti-inflammatory mechanisms and significance of SGLT2 inhibition in HF patients remain to be determined and appear promising.

Development of therapies targeting immunometabolism is still in the early phases, partly due to the challenges of specifically targeting certain immune cell populations while avoiding other tissues.^{177, 178} Most immunometabolic studies have focused on targeting tumor-associated macrophages in cancer, or aberrant B and T lymphocyte function during autoimmune disease.^{156, 178} Thus, immunometabolic drugs may need to be disease-specific. The anti-inflammatory drug dimethyl fumarate (DMF) was recently found to inhibit GAPDH specifically in activated myeloid cells, leading to inhibition of glycolysis and decreasing inflammation in a mouse model of multiple sclerosis.¹⁷⁹ DMF is clinically approved for treating inflammation in psoriasis and has cardioprotective properties in rodent models, but whether its cardioprotective effects are dependent on monocytes/macrophages is unknown.¹⁸⁰ Thus, while exploration of new therapies targeting immunometabolism is warranted for treatment of CVD and HF, some of the existing therapies may act through previously unidentified metabolic pathways.

Role of Neutrophils and Lymphocytes

Macrophages display high metabolic adaptability relative to other immune cells, and can quickly undergo metabolic shifts based on their environment. For this reason, immunometabolic studies have mainly focused on macrophages.¹⁸¹ However, metabolic reprogramming may also underlie T lymphocyte function during cardiac injury.¹⁸² Both CD4+ and CD8+ T cells have been implicated in cardiac remodeling, and play crucial roles in regulating macrophage functions after MI via secretion of pro-inflammatory cytokines (interferon-gamma) and anti-inflammatory cytokines (IL-4/IL-13).^{21, 182, 183} In autoimmune-induced dilated cardiomyopathy, CD4+ T cells show enhanced glycolytic metabolism, the inhibition of which prevents cardiac injury.¹⁸²

Neutrophils are another immune cell that participate in cardiac remodeling, particularly in the early inflammatory phase.¹⁵⁷ While circulating neutrophils have been observed to be primarily glycolytic, their metabolic adaptations and roles in the injured heart remain to be characterized.¹⁵⁷ Thus, immunometabolic therapies must be considered in the context of both innate and adaptive immune cells.

CONCLUSIONS AND PERSPECTIVES

Obesity and HTN often coexist and are characterized by metabolic dysfunction and inflammation of the heart. While inflammation is recognized to adversely affect cardiac metabolism (e.g. exacerbation of insulin resistance), the mechanisms by which metabolic changes affect cardiac macrophage function are poorly understood. Obesity and HTN both lead to cardiac injury and remodeling through overlapping mechanisms including mechanical stress, neurohormonal activation, metabolic derangements, and as we have discussed, macrophage-mediated inflammation. Likewise, obesity and HTN activation of pro-inflammatory M1 macrophage polarization reinforces this vicious cycle. While several mechanisms may mediate M1 polarization in the metabolic syndrome, there is substantial evidence that alterations in macrophage metabolism due to changes in metabolic substrates and pro-inflammatory lipids play a critical role, priming macrophages to promote adverse outcomes as HTN develops.

Obesity is a major risk factor for HTN, and exacerbates target-organ injury and HTN complications. Thus, it remains critically important to understand the mechanisms by which obesity and HTN synergistically promote cardiac injury, and to consider the obesity-HTN spectrum as a continuous rather than separate entity, particularly in regards to inflammation. We have discussed the major findings in animal models, however, it is important to note that many of these studies show obesity and HTN in isolation, whereas in humans obesity and HTN are often coexist since obesity accounts for 65–75 % of the risk for primary HTN.⁸ Thus, better animal models are needed to recapitulate the human condition.

While inflammation is a promising target, anti-inflammatory therapies are still in development, and further studies need to evaluate the potential of altering macrophage metabolism. The challenges in understanding and targeting cardiac immunometabolism will include 1) macrophage heterogeneity in the remodeling heart, 2) different macrophage roles and mechanisms associated with systolic and diastolic dysfunction,²⁴ and 3) directly targeting macrophage metabolism without adversely affecting cardiomyocyte metabolism. Cardiac immunometabolism is still a young field, and shows promise given the importance of systemic and cardiac metabolism and inflammation in driving HF progression.

LIST OF ABBREVIATIONS

2-DG

2-deoxyglucose

AGE

advanced glycation end-product

Ang II

angiotensin II

CAD

coronary artery disease

CCR2

C-C chemokine receptor 2

cGMP

cyclic guanosine monophosphate

CKD

chronic kidney disease

COX

cyclooxygenase

CVD

cardiovascular disease

DAMP

damage-associated molecular pattern

DMF

dimethyl fumarate

EAT

epicardial adipose tissue

ECM

extracellular matrix

ETC

electron transport chain

FAO

fatty acid oxidation

GAIT

gamma-interferon- activated inhibitor of translation

GAPDH

glyceraldehyde-3- phosphate dehydrogenase

GLUT1

glucose transporter 1

GPCR

G-protein coupled receptor

HF

heart failure

HFpEF

heart failure with preserved ejection fraction

HFrEF

heart failure with reduced ejection fraction

HIF-1α

hypoxia-inducible factor-1 alpha

HRE

hypoxia response element

HTN

hypertension

IL-1β

interleukin-1 beta

iNOS

inducible nitric oxide

LOX

lipoxygenase

LV

left ventricle

MI

myocardial infarction

MMP

matrix metalloproteinase

NF-κB

nuclear factor kappa-light- chain-enhancer of activated B cells

NO

nitric oxide

OXPHOS

oxidative phosphorylation

PDE

phosphodiesterase

PDK1

pyruvate dehydrogenase kinase 1

PGC-1β

PPAR-γ coactivator-1β

PKG

protein kinase G

PPAR-γ

Peroxisome proliferator activated receptor-gamma

RAAS

renin-angiotensin aldosterone system

ROS

reactive oxygen species

RyR

ryanodine receptor

SDH

succinate dehydrogenase

SERCA2a

sarcoplasmic reticulum Ca²⁺ ATPase

SGK1

serum glucocorticoid kinase

SGLT2

sodium glucose transporter 2

SNS

sympathetic nervous system

SR

sarcoplasmic reticulum

STAT6

signal transducer and activator of transcription 6

TCA

tricarboxylic acid cycle

TGF-β1

transforming growth factor beta-1

TLR4

toll-like receptor 4

TMZ

trimetazidine

TNF-α

tumor necrosis factor alpha

TZD

thiazolidinedione

UCP-2

uncoupling protein 2

VEGF

vascular endothelial growth factor